You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@myfaces.apache.org by "Stevan Malesevic (JIRA)" <de...@myfaces.apache.org> on 2010/12/20 16:46:01 UTC
[jira] Created: (TRINIDAD-1985) High live memory usage from
SkinStyleProvider
High live memory usage from SkinStyleProvider
---------------------------------------------
Key: TRINIDAD-1985
URL: https://issues.apache.org/jira/browse/TRINIDAD-1985
Project: MyFaces Trinidad
Issue Type: Bug
Components: Archetype
Affects Versions: 1.2.12-core
Reporter: Stevan Malesevic
We were checking a live memory usage in one app and noticed that some 83MB of heap is pinned under SkinStyleProvider. This is under 14 instances of FileSystemStyleCache$Entry each with about 6MB of memory
Heap shows these keys for styles
Locale Version
en Mozilla/5.0 (Windows NT 5.1; rv:2.0b7) Gecko/20100101 Firefox/4.0b7
ko Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US; rv:1.9.1.11) Gecko/20100701 Firefox/3.5.11 ( .NET CLR 3.5.30729)
en Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US; rv:1.9.1.11) Gecko/20100701 Firefox/3.5.11 ( .NET CLR 3.5.30729)
ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/534.10 (KHTML, like Gecko) Chrome/8.0.552.224 Safari/534.10
en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/534.10 (KHTML, like Gecko) Chrome/8.0.552.224 Safari/534.10
ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.3) Gecko/20090824 Firefox/3.5.3
en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.3) Gecko/20090824 Firefox/3.5.3
en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.0.18) Gecko/2010020220 Firefox/3.0.18 (.NET CLR 3.5.30729)
ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.0.18) Gecko/2010020220 Firefox/3.0.18 (.NET CLR 3.5.30729)
en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.9) Gecko/20100315 Firefox/3.5.9 (.NET CLR 3.5.30729)
ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.9) Gecko/20100315 Firefox/3.5.9 (.NET CLR 3.5.30729)
en Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 5.2; Trident/4.0; .NET CLR 1.1.4322; .NET CLR 2.0.50727; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729)
en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2) Gecko/20100115 Firefox/3.6 (.NET CLR 3.5.30729)
en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.15) Gecko/20101026 Firefox/3.5.15 ( .NET CLR 3.5.30729)
en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.10) Gecko/20100914 Firefox/3.6.10 (.NET CLR 3.5.30729)
ko Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; .NET CLR 2.0.50727; .NET CLR 1.1.4322; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; InfoPath.2)
en Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; .NET CLR 2.0.50727; .NET CLR 1.1.4322; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; InfoPath.2)
en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.12) Gecko/20101026 Firefox/3.6.12 ( .NET CLR 3.5.30729)
en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.16) Gecko/20101130 Firefox/3.5.16 ( .NET CLR 3.5.30729)
ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.16) Gecko/20101130 Firefox/3.5.16 ( .NET CLR 3.5.30729)
en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.13) Gecko/20101203 Firefox/3.6.13
Now beside the amount of memory pinned by single FileSystemStyleCache$Entry the issue is that we apparently create too many different versions and there is no restriction to the size of cache leaving open door for memory leak
Two questions
1. Do we really have different styles for FF on 3.5 or 3.6 or so? If not why do we key by it?
2. Given different browsers and locales having cache as plain concurrent hash map is actually a source of leak as over time it can accumulate more and more. Do we have any restriction there? Should the entries by LRU or have exparation
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (TRINIDAD-1985) High live memory usage from
SkinStyleProvider
Posted by "Jeanne Waldman (JIRA)" <de...@myfaces.apache.org>.
[ https://issues.apache.org/jira/browse/TRINIDAD-1985?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12977569#action_12977569 ]
Jeanne Waldman commented on TRINIDAD-1985:
------------------------------------------
For me to answer #1, I would need to know what skin you were using, and then I would look at the CSS files for that skin to see if we have different css for FF on 3.5 and 3.6 (like @agent gecko blocks).
What skin were you using?
We have two types of caches in FileSystemStyleCache. One is a derivationKey, and that keys off of the styleSheetNodes, so even if you are using different browsers or locales, if they have the same styleSheetNodes, then you will get the same cached value.
> High live memory usage from SkinStyleProvider
> ---------------------------------------------
>
> Key: TRINIDAD-1985
> URL: https://issues.apache.org/jira/browse/TRINIDAD-1985
> Project: MyFaces Trinidad
> Issue Type: Bug
> Components: Archetype
> Affects Versions: 1.2.12-core
> Reporter: Stevan Malesevic
>
> We were checking a live memory usage in one app and noticed that some 83MB of heap is pinned under SkinStyleProvider. This is under 14 instances of FileSystemStyleCache$Entry each with about 6MB of memory
> Heap shows these keys for styles
> Locale Version
> en Mozilla/5.0 (Windows NT 5.1; rv:2.0b7) Gecko/20100101 Firefox/4.0b7
> ko Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US; rv:1.9.1.11) Gecko/20100701 Firefox/3.5.11 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US; rv:1.9.1.11) Gecko/20100701 Firefox/3.5.11 ( .NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/534.10 (KHTML, like Gecko) Chrome/8.0.552.224 Safari/534.10
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/534.10 (KHTML, like Gecko) Chrome/8.0.552.224 Safari/534.10
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.3) Gecko/20090824 Firefox/3.5.3
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.3) Gecko/20090824 Firefox/3.5.3
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.0.18) Gecko/2010020220 Firefox/3.0.18 (.NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.0.18) Gecko/2010020220 Firefox/3.0.18 (.NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.9) Gecko/20100315 Firefox/3.5.9 (.NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.9) Gecko/20100315 Firefox/3.5.9 (.NET CLR 3.5.30729)
> en Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 5.2; Trident/4.0; .NET CLR 1.1.4322; .NET CLR 2.0.50727; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2) Gecko/20100115 Firefox/3.6 (.NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.15) Gecko/20101026 Firefox/3.5.15 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.10) Gecko/20100914 Firefox/3.6.10 (.NET CLR 3.5.30729)
> ko Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; .NET CLR 2.0.50727; .NET CLR 1.1.4322; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; InfoPath.2)
> en Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; .NET CLR 2.0.50727; .NET CLR 1.1.4322; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; InfoPath.2)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.12) Gecko/20101026 Firefox/3.6.12 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.16) Gecko/20101130 Firefox/3.5.16 ( .NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.16) Gecko/20101130 Firefox/3.5.16 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.13) Gecko/20101203 Firefox/3.6.13
> Now beside the amount of memory pinned by single FileSystemStyleCache$Entry the issue is that we apparently create too many different versions and there is no restriction to the size of cache leaving open door for memory leak
> Two questions
> 1. Do we really have different styles for FF on 3.5 or 3.6 or so? If not why do we key by it?
> 2. Given different browsers and locales having cache as plain concurrent hash map is actually a source of leak as over time it can accumulate more and more. Do we have any restriction there? Should the entries by LRU or have exparation
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (TRINIDAD-1985) High live memory usage from
SkinStyleProvider
Posted by "Jeanne Waldman (JIRA)" <de...@myfaces.apache.org>.
[ https://issues.apache.org/jira/browse/TRINIDAD-1985?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12981069#action_12981069 ]
Jeanne Waldman commented on TRINIDAD-1985:
------------------------------------------
A suggestion from Blake - look into using a soft reference instead of a LRUCache.
> High live memory usage from SkinStyleProvider
> ---------------------------------------------
>
> Key: TRINIDAD-1985
> URL: https://issues.apache.org/jira/browse/TRINIDAD-1985
> Project: MyFaces Trinidad
> Issue Type: Bug
> Components: Archetype
> Affects Versions: 1.2.12-core
> Reporter: Stevan Malesevic
>
> We were checking a live memory usage in one app and noticed that some 83MB of heap is pinned under SkinStyleProvider. This is under 14 instances of FileSystemStyleCache$Entry each with about 6MB of memory
> Heap shows these keys for styles
> Locale Version
> en Mozilla/5.0 (Windows NT 5.1; rv:2.0b7) Gecko/20100101 Firefox/4.0b7
> ko Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US; rv:1.9.1.11) Gecko/20100701 Firefox/3.5.11 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US; rv:1.9.1.11) Gecko/20100701 Firefox/3.5.11 ( .NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/534.10 (KHTML, like Gecko) Chrome/8.0.552.224 Safari/534.10
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/534.10 (KHTML, like Gecko) Chrome/8.0.552.224 Safari/534.10
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.3) Gecko/20090824 Firefox/3.5.3
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.3) Gecko/20090824 Firefox/3.5.3
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.0.18) Gecko/2010020220 Firefox/3.0.18 (.NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.0.18) Gecko/2010020220 Firefox/3.0.18 (.NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.9) Gecko/20100315 Firefox/3.5.9 (.NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.9) Gecko/20100315 Firefox/3.5.9 (.NET CLR 3.5.30729)
> en Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 5.2; Trident/4.0; .NET CLR 1.1.4322; .NET CLR 2.0.50727; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2) Gecko/20100115 Firefox/3.6 (.NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.15) Gecko/20101026 Firefox/3.5.15 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.10) Gecko/20100914 Firefox/3.6.10 (.NET CLR 3.5.30729)
> ko Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; .NET CLR 2.0.50727; .NET CLR 1.1.4322; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; InfoPath.2)
> en Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; .NET CLR 2.0.50727; .NET CLR 1.1.4322; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; InfoPath.2)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.12) Gecko/20101026 Firefox/3.6.12 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.16) Gecko/20101130 Firefox/3.5.16 ( .NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.16) Gecko/20101130 Firefox/3.5.16 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.13) Gecko/20101203 Firefox/3.6.13
> Now beside the amount of memory pinned by single FileSystemStyleCache$Entry the issue is that we apparently create too many different versions and there is no restriction to the size of cache leaving open door for memory leak
> Two questions
> 1. Do we really have different styles for FF on 3.5 or 3.6 or so? If not why do we key by it?
> 2. Given different browsers and locales having cache as plain concurrent hash map is actually a source of leak as over time it can accumulate more and more. Do we have any restriction there? Should the entries by LRU or have exparation
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (TRINIDAD-1985) High live memory usage from
SkinStyleProvider
Posted by "Jeanne Waldman (JIRA)" <de...@myfaces.apache.org>.
[ https://issues.apache.org/jira/browse/TRINIDAD-1985?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12986088#action_12986088 ]
Jeanne Waldman commented on TRINIDAD-1985:
------------------------------------------
I reworked the code to reuse the CSSStyle objects at the FileSystemStyleCache level, not the StylesImpl level, and this looks like it saves a lot of memory. I'll upload a patch once more tests have been done.
> High live memory usage from SkinStyleProvider
> ---------------------------------------------
>
> Key: TRINIDAD-1985
> URL: https://issues.apache.org/jira/browse/TRINIDAD-1985
> Project: MyFaces Trinidad
> Issue Type: Bug
> Components: Archetype
> Affects Versions: 1.2.12-core
> Reporter: Stevan Malesevic
>
> We were checking a live memory usage in one app and noticed that some 83MB of heap is pinned under SkinStyleProvider. This is under 14 instances of FileSystemStyleCache$Entry each with about 6MB of memory
> Heap shows these keys for styles
> Locale Version
> en Mozilla/5.0 (Windows NT 5.1; rv:2.0b7) Gecko/20100101 Firefox/4.0b7
> ko Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US; rv:1.9.1.11) Gecko/20100701 Firefox/3.5.11 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US; rv:1.9.1.11) Gecko/20100701 Firefox/3.5.11 ( .NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/534.10 (KHTML, like Gecko) Chrome/8.0.552.224 Safari/534.10
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/534.10 (KHTML, like Gecko) Chrome/8.0.552.224 Safari/534.10
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.3) Gecko/20090824 Firefox/3.5.3
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.3) Gecko/20090824 Firefox/3.5.3
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.0.18) Gecko/2010020220 Firefox/3.0.18 (.NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.0.18) Gecko/2010020220 Firefox/3.0.18 (.NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.9) Gecko/20100315 Firefox/3.5.9 (.NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.9) Gecko/20100315 Firefox/3.5.9 (.NET CLR 3.5.30729)
> en Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 5.2; Trident/4.0; .NET CLR 1.1.4322; .NET CLR 2.0.50727; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2) Gecko/20100115 Firefox/3.6 (.NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.15) Gecko/20101026 Firefox/3.5.15 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.10) Gecko/20100914 Firefox/3.6.10 (.NET CLR 3.5.30729)
> ko Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; .NET CLR 2.0.50727; .NET CLR 1.1.4322; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; InfoPath.2)
> en Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; .NET CLR 2.0.50727; .NET CLR 1.1.4322; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; InfoPath.2)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.12) Gecko/20101026 Firefox/3.6.12 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.16) Gecko/20101130 Firefox/3.5.16 ( .NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.16) Gecko/20101130 Firefox/3.5.16 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.13) Gecko/20101203 Firefox/3.6.13
> Now beside the amount of memory pinned by single FileSystemStyleCache$Entry the issue is that we apparently create too many different versions and there is no restriction to the size of cache leaving open door for memory leak
> Two questions
> 1. Do we really have different styles for FF on 3.5 or 3.6 or so? If not why do we key by it?
> 2. Given different browsers and locales having cache as plain concurrent hash map is actually a source of leak as over time it can accumulate more and more. Do we have any restriction there? Should the entries by LRU or have exparation
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (TRINIDAD-1985) High live memory usage from
SkinStyleProvider
Posted by "Jeanne Waldman (JIRA)" <de...@myfaces.apache.org>.
[ https://issues.apache.org/jira/browse/TRINIDAD-1985?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12978467#action_12978467 ]
Jeanne Waldman commented on TRINIDAD-1985:
------------------------------------------
While investigating this issue, I found that if we have:
@agent ie, gecko
{
.fooJMWTestSSN {background-color:pink}
}
the CSS file is shared if this is the only @rule.
If I have:
@agent ie
{
.fooJMWTestSSN {background-color:pink}
}
@agent gecko
{
.fooJMWTestSSN {background-color:pink}
}
then this creates two different css files (one for gecko and one for IE) even though there are no diffs in the contents of the css file.
A simple performance enhancement is to make sure the @rules that can be combined, are.
A more complicated enhancement is to do a better job of comparing stylesheet nodes.
> High live memory usage from SkinStyleProvider
> ---------------------------------------------
>
> Key: TRINIDAD-1985
> URL: https://issues.apache.org/jira/browse/TRINIDAD-1985
> Project: MyFaces Trinidad
> Issue Type: Bug
> Components: Archetype
> Affects Versions: 1.2.12-core
> Reporter: Stevan Malesevic
>
> We were checking a live memory usage in one app and noticed that some 83MB of heap is pinned under SkinStyleProvider. This is under 14 instances of FileSystemStyleCache$Entry each with about 6MB of memory
> Heap shows these keys for styles
> Locale Version
> en Mozilla/5.0 (Windows NT 5.1; rv:2.0b7) Gecko/20100101 Firefox/4.0b7
> ko Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US; rv:1.9.1.11) Gecko/20100701 Firefox/3.5.11 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US; rv:1.9.1.11) Gecko/20100701 Firefox/3.5.11 ( .NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/534.10 (KHTML, like Gecko) Chrome/8.0.552.224 Safari/534.10
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/534.10 (KHTML, like Gecko) Chrome/8.0.552.224 Safari/534.10
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.3) Gecko/20090824 Firefox/3.5.3
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.3) Gecko/20090824 Firefox/3.5.3
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.0.18) Gecko/2010020220 Firefox/3.0.18 (.NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.0.18) Gecko/2010020220 Firefox/3.0.18 (.NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.9) Gecko/20100315 Firefox/3.5.9 (.NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.9) Gecko/20100315 Firefox/3.5.9 (.NET CLR 3.5.30729)
> en Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 5.2; Trident/4.0; .NET CLR 1.1.4322; .NET CLR 2.0.50727; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2) Gecko/20100115 Firefox/3.6 (.NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.15) Gecko/20101026 Firefox/3.5.15 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.10) Gecko/20100914 Firefox/3.6.10 (.NET CLR 3.5.30729)
> ko Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; .NET CLR 2.0.50727; .NET CLR 1.1.4322; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; InfoPath.2)
> en Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; .NET CLR 2.0.50727; .NET CLR 1.1.4322; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; InfoPath.2)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.12) Gecko/20101026 Firefox/3.6.12 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.16) Gecko/20101130 Firefox/3.5.16 ( .NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.16) Gecko/20101130 Firefox/3.5.16 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.13) Gecko/20101203 Firefox/3.6.13
> Now beside the amount of memory pinned by single FileSystemStyleCache$Entry the issue is that we apparently create too many different versions and there is no restriction to the size of cache leaving open door for memory leak
> Two questions
> 1. Do we really have different styles for FF on 3.5 or 3.6 or so? If not why do we key by it?
> 2. Given different browsers and locales having cache as plain concurrent hash map is actually a source of leak as over time it can accumulate more and more. Do we have any restriction there? Should the entries by LRU or have exparation
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Issue Comment Edited: (TRINIDAD-1985) High live memory usage
from SkinStyleProvider
Posted by "Jeanne Waldman (JIRA)" <de...@myfaces.apache.org>.
[ https://issues.apache.org/jira/browse/TRINIDAD-1985?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12980516#action_12980516 ]
Jeanne Waldman edited comment on TRINIDAD-1985 at 1/11/11 11:08 PM:
--------------------------------------------------------------------
Something to think about --
In the FileSystemStyleCache code, the _cache and _entryCache are ConcurrentMaps. If we use an LRU implementation org.apache.myfaces.trinidadinternal.util.LRUCache extends LinkedHashMap, and wrap this in Collections.synchronizedMap, then we will lose the features of ConcurrentMap; we will lose scalability since only one thread can access the Map at once with synchronizedMap.
was (Author: jeanne.waldman@oracle.com):
Something to think about --
In the FileSystemStyleCache code, the _cache and _entryCache are ConcurrentMaps. If we use an LRU implementation org.apache.myfaces.trinidadinternal.util.LRUCache extends LinkedHashMap, and wrap this in Collections.synchronizedMap, then we will lose the features of ConcurrentMap. We'll need to synchronize on remove, and from what I've read, we will lose scalability since only one thread can access the Map at once with synchronizedMap.
> High live memory usage from SkinStyleProvider
> ---------------------------------------------
>
> Key: TRINIDAD-1985
> URL: https://issues.apache.org/jira/browse/TRINIDAD-1985
> Project: MyFaces Trinidad
> Issue Type: Bug
> Components: Archetype
> Affects Versions: 1.2.12-core
> Reporter: Stevan Malesevic
>
> We were checking a live memory usage in one app and noticed that some 83MB of heap is pinned under SkinStyleProvider. This is under 14 instances of FileSystemStyleCache$Entry each with about 6MB of memory
> Heap shows these keys for styles
> Locale Version
> en Mozilla/5.0 (Windows NT 5.1; rv:2.0b7) Gecko/20100101 Firefox/4.0b7
> ko Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US; rv:1.9.1.11) Gecko/20100701 Firefox/3.5.11 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US; rv:1.9.1.11) Gecko/20100701 Firefox/3.5.11 ( .NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/534.10 (KHTML, like Gecko) Chrome/8.0.552.224 Safari/534.10
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/534.10 (KHTML, like Gecko) Chrome/8.0.552.224 Safari/534.10
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.3) Gecko/20090824 Firefox/3.5.3
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.3) Gecko/20090824 Firefox/3.5.3
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.0.18) Gecko/2010020220 Firefox/3.0.18 (.NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.0.18) Gecko/2010020220 Firefox/3.0.18 (.NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.9) Gecko/20100315 Firefox/3.5.9 (.NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.9) Gecko/20100315 Firefox/3.5.9 (.NET CLR 3.5.30729)
> en Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 5.2; Trident/4.0; .NET CLR 1.1.4322; .NET CLR 2.0.50727; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2) Gecko/20100115 Firefox/3.6 (.NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.15) Gecko/20101026 Firefox/3.5.15 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.10) Gecko/20100914 Firefox/3.6.10 (.NET CLR 3.5.30729)
> ko Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; .NET CLR 2.0.50727; .NET CLR 1.1.4322; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; InfoPath.2)
> en Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; .NET CLR 2.0.50727; .NET CLR 1.1.4322; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; InfoPath.2)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.12) Gecko/20101026 Firefox/3.6.12 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.16) Gecko/20101130 Firefox/3.5.16 ( .NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.16) Gecko/20101130 Firefox/3.5.16 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.13) Gecko/20101203 Firefox/3.6.13
> Now beside the amount of memory pinned by single FileSystemStyleCache$Entry the issue is that we apparently create too many different versions and there is no restriction to the size of cache leaving open door for memory leak
> Two questions
> 1. Do we really have different styles for FF on 3.5 or 3.6 or so? If not why do we key by it?
> 2. Given different browsers and locales having cache as plain concurrent hash map is actually a source of leak as over time it can accumulate more and more. Do we have any restriction there? Should the entries by LRU or have exparation
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Issue Comment Edited: (TRINIDAD-1985) High live memory usage
from SkinStyleProvider
Posted by "Jeanne Waldman (JIRA)" <de...@myfaces.apache.org>.
[ https://issues.apache.org/jira/browse/TRINIDAD-1985?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12977569#action_12977569 ]
Jeanne Waldman edited comment on TRINIDAD-1985 at 1/4/11 9:06 PM:
------------------------------------------------------------------
For me to answer #1, I would need to know what skin you were using, and then I would look at the CSS files for that skin to see if we have different css for FF on 3.5 and 3.6 (like @agent gecko blocks).
I see in the adf faces skins, we do have different styles for gecko 1.9.2 and lower:
/* Firefox versions earlier than 3.6 (Gecko 1.9.2.0) do no support multiple background images */
And we do have slightly different styles for locales in Trinidad's base-desktop.css. We might be able to fix this if we can get rid of DefaultServerFont:alias. Maybe this isn't being used.
We have two types of caches in FileSystemStyleCache. One is am exact match (browser, locale, etc) key, and another is the derivationKey, and that keys off of the styleSheetNodes, so even if you are using different browsers or locales, if they have the same styleSheetNodes, then you will get the same cached value.
What skin were you using?
If we go with a most recently used limit, what do you think a reasonable limit would be?
was (Author: jeanne.waldman@oracle.com):
For me to answer #1, I would need to know what skin you were using, and then I would look at the CSS files for that skin to see if we have different css for FF on 3.5 and 3.6 (like @agent gecko blocks).
I see in the adf faces skins, we do have different styles for gecko 1.9.2 and lower:
/* Firefox versions earlier than 3.6 (Gecko 1.9.2.0) do no support multiple background images */
And we do have slightly different styles for ko in Trinidad's base-desktop.css
We have two types of caches in FileSystemStyleCache. One is am exact match (browser, locale, etc) key, and another is the derivationKey, and that keys off of the styleSheetNodes, so even if you are using different browsers or locales, if they have the same styleSheetNodes, then you will get the same cached value.
What skin were you using?
If we go with a most recently used limit, what do you think a reasonable limit would be?
> High live memory usage from SkinStyleProvider
> ---------------------------------------------
>
> Key: TRINIDAD-1985
> URL: https://issues.apache.org/jira/browse/TRINIDAD-1985
> Project: MyFaces Trinidad
> Issue Type: Bug
> Components: Archetype
> Affects Versions: 1.2.12-core
> Reporter: Stevan Malesevic
>
> We were checking a live memory usage in one app and noticed that some 83MB of heap is pinned under SkinStyleProvider. This is under 14 instances of FileSystemStyleCache$Entry each with about 6MB of memory
> Heap shows these keys for styles
> Locale Version
> en Mozilla/5.0 (Windows NT 5.1; rv:2.0b7) Gecko/20100101 Firefox/4.0b7
> ko Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US; rv:1.9.1.11) Gecko/20100701 Firefox/3.5.11 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US; rv:1.9.1.11) Gecko/20100701 Firefox/3.5.11 ( .NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/534.10 (KHTML, like Gecko) Chrome/8.0.552.224 Safari/534.10
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/534.10 (KHTML, like Gecko) Chrome/8.0.552.224 Safari/534.10
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.3) Gecko/20090824 Firefox/3.5.3
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.3) Gecko/20090824 Firefox/3.5.3
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.0.18) Gecko/2010020220 Firefox/3.0.18 (.NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.0.18) Gecko/2010020220 Firefox/3.0.18 (.NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.9) Gecko/20100315 Firefox/3.5.9 (.NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.9) Gecko/20100315 Firefox/3.5.9 (.NET CLR 3.5.30729)
> en Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 5.2; Trident/4.0; .NET CLR 1.1.4322; .NET CLR 2.0.50727; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2) Gecko/20100115 Firefox/3.6 (.NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.15) Gecko/20101026 Firefox/3.5.15 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.10) Gecko/20100914 Firefox/3.6.10 (.NET CLR 3.5.30729)
> ko Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; .NET CLR 2.0.50727; .NET CLR 1.1.4322; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; InfoPath.2)
> en Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; .NET CLR 2.0.50727; .NET CLR 1.1.4322; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; InfoPath.2)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.12) Gecko/20101026 Firefox/3.6.12 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.16) Gecko/20101130 Firefox/3.5.16 ( .NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.16) Gecko/20101130 Firefox/3.5.16 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.13) Gecko/20101203 Firefox/3.6.13
> Now beside the amount of memory pinned by single FileSystemStyleCache$Entry the issue is that we apparently create too many different versions and there is no restriction to the size of cache leaving open door for memory leak
> Two questions
> 1. Do we really have different styles for FF on 3.5 or 3.6 or so? If not why do we key by it?
> 2. Given different browsers and locales having cache as plain concurrent hash map is actually a source of leak as over time it can accumulate more and more. Do we have any restriction there? Should the entries by LRU or have exparation
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (TRINIDAD-1985) High live memory usage from
SkinStyleProvider
Posted by "Jeanne Waldman (JIRA)" <de...@myfaces.apache.org>.
[ https://issues.apache.org/jira/browse/TRINIDAD-1985?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12989910#comment-12989910 ]
Jeanne Waldman commented on TRINIDAD-1985:
------------------------------------------
I fixed TRINIDAD-2025 move _styleNodeToStyleMap to FileSystemStyleCache to take advantage of same entires on the higher, browser or version or platform layer
which brings down the memory consumption by about 50%.
If we limit it to 40 entries, what is the likelihood that we would get up to 40 entries? Does it matter as much now that we reduced the memory by so much?
related issue TRINIDAD-2026 memory leak in skinstyleprovider. This occurs if one application uses a bunch of different skins.
> High live memory usage from SkinStyleProvider
> ---------------------------------------------
>
> Key: TRINIDAD-1985
> URL: https://issues.apache.org/jira/browse/TRINIDAD-1985
> Project: MyFaces Trinidad
> Issue Type: Bug
> Components: Archetype
> Affects Versions: 1.2.12-core
> Reporter: Stevan Malesevic
> Assignee: Jeanne Waldman
>
> We were checking a live memory usage in one app and noticed that some 83MB of heap is pinned under SkinStyleProvider. This is under 14 instances of FileSystemStyleCache$Entry each with about 6MB of memory
> Heap shows these keys for styles
> Locale Version
> en Mozilla/5.0 (Windows NT 5.1; rv:2.0b7) Gecko/20100101 Firefox/4.0b7
> ko Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US; rv:1.9.1.11) Gecko/20100701 Firefox/3.5.11 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US; rv:1.9.1.11) Gecko/20100701 Firefox/3.5.11 ( .NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/534.10 (KHTML, like Gecko) Chrome/8.0.552.224 Safari/534.10
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/534.10 (KHTML, like Gecko) Chrome/8.0.552.224 Safari/534.10
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.3) Gecko/20090824 Firefox/3.5.3
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.3) Gecko/20090824 Firefox/3.5.3
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.0.18) Gecko/2010020220 Firefox/3.0.18 (.NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.0.18) Gecko/2010020220 Firefox/3.0.18 (.NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.9) Gecko/20100315 Firefox/3.5.9 (.NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.9) Gecko/20100315 Firefox/3.5.9 (.NET CLR 3.5.30729)
> en Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 5.2; Trident/4.0; .NET CLR 1.1.4322; .NET CLR 2.0.50727; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2) Gecko/20100115 Firefox/3.6 (.NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.15) Gecko/20101026 Firefox/3.5.15 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.10) Gecko/20100914 Firefox/3.6.10 (.NET CLR 3.5.30729)
> ko Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; .NET CLR 2.0.50727; .NET CLR 1.1.4322; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; InfoPath.2)
> en Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; .NET CLR 2.0.50727; .NET CLR 1.1.4322; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; InfoPath.2)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.12) Gecko/20101026 Firefox/3.6.12 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.16) Gecko/20101130 Firefox/3.5.16 ( .NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.16) Gecko/20101130 Firefox/3.5.16 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.13) Gecko/20101203 Firefox/3.6.13
> Now beside the amount of memory pinned by single FileSystemStyleCache$Entry the issue is that we apparently create too many different versions and there is no restriction to the size of cache leaving open door for memory leak
> Two questions
> 1. Do we really have different styles for FF on 3.5 or 3.6 or so? If not why do we key by it?
> 2. Given different browsers and locales having cache as plain concurrent hash map is actually a source of leak as over time it can accumulate more and more. Do we have any restriction there? Should the entries by LRU or have exparation
--
This message is automatically generated by JIRA.
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] Commented: (TRINIDAD-1985) High live memory usage from
SkinStyleProvider
Posted by "Jeanne Waldman (JIRA)" <de...@myfaces.apache.org>.
[ https://issues.apache.org/jira/browse/TRINIDAD-1985?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12981976#action_12981976 ]
Jeanne Waldman commented on TRINIDAD-1985:
------------------------------------------
The soft references and LRU caching would help clean up rarely used css files but if the users are actually using all of these pretty consistently, the soft references aren't going to help much and could actually make things worse by thrashing as the skin information gets purged and recreated constantly if the application is under memory pressure.
Look into re-using the CSSStyle objects more efficiently as a workaround.
> High live memory usage from SkinStyleProvider
> ---------------------------------------------
>
> Key: TRINIDAD-1985
> URL: https://issues.apache.org/jira/browse/TRINIDAD-1985
> Project: MyFaces Trinidad
> Issue Type: Bug
> Components: Archetype
> Affects Versions: 1.2.12-core
> Reporter: Stevan Malesevic
>
> We were checking a live memory usage in one app and noticed that some 83MB of heap is pinned under SkinStyleProvider. This is under 14 instances of FileSystemStyleCache$Entry each with about 6MB of memory
> Heap shows these keys for styles
> Locale Version
> en Mozilla/5.0 (Windows NT 5.1; rv:2.0b7) Gecko/20100101 Firefox/4.0b7
> ko Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US; rv:1.9.1.11) Gecko/20100701 Firefox/3.5.11 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US; rv:1.9.1.11) Gecko/20100701 Firefox/3.5.11 ( .NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/534.10 (KHTML, like Gecko) Chrome/8.0.552.224 Safari/534.10
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/534.10 (KHTML, like Gecko) Chrome/8.0.552.224 Safari/534.10
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.3) Gecko/20090824 Firefox/3.5.3
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.3) Gecko/20090824 Firefox/3.5.3
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.0.18) Gecko/2010020220 Firefox/3.0.18 (.NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.0.18) Gecko/2010020220 Firefox/3.0.18 (.NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.9) Gecko/20100315 Firefox/3.5.9 (.NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.9) Gecko/20100315 Firefox/3.5.9 (.NET CLR 3.5.30729)
> en Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 5.2; Trident/4.0; .NET CLR 1.1.4322; .NET CLR 2.0.50727; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2) Gecko/20100115 Firefox/3.6 (.NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.15) Gecko/20101026 Firefox/3.5.15 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.10) Gecko/20100914 Firefox/3.6.10 (.NET CLR 3.5.30729)
> ko Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; .NET CLR 2.0.50727; .NET CLR 1.1.4322; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; InfoPath.2)
> en Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; .NET CLR 2.0.50727; .NET CLR 1.1.4322; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; InfoPath.2)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.12) Gecko/20101026 Firefox/3.6.12 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.16) Gecko/20101130 Firefox/3.5.16 ( .NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.16) Gecko/20101130 Firefox/3.5.16 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.13) Gecko/20101203 Firefox/3.6.13
> Now beside the amount of memory pinned by single FileSystemStyleCache$Entry the issue is that we apparently create too many different versions and there is no restriction to the size of cache leaving open door for memory leak
> Two questions
> 1. Do we really have different styles for FF on 3.5 or 3.6 or so? If not why do we key by it?
> 2. Given different browsers and locales having cache as plain concurrent hash map is actually a source of leak as over time it can accumulate more and more. Do we have any restriction there? Should the entries by LRU or have exparation
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (TRINIDAD-1985) High live memory usage from
SkinStyleProvider
Posted by "Jeanne Waldman (JIRA)" <de...@myfaces.apache.org>.
[ https://issues.apache.org/jira/browse/TRINIDAD-1985?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12977574#action_12977574 ]
Jeanne Waldman commented on TRINIDAD-1985:
------------------------------------------
One more question.
Were you using one skin but with different browsers/locales or 14 different skins or something in between?
> High live memory usage from SkinStyleProvider
> ---------------------------------------------
>
> Key: TRINIDAD-1985
> URL: https://issues.apache.org/jira/browse/TRINIDAD-1985
> Project: MyFaces Trinidad
> Issue Type: Bug
> Components: Archetype
> Affects Versions: 1.2.12-core
> Reporter: Stevan Malesevic
>
> We were checking a live memory usage in one app and noticed that some 83MB of heap is pinned under SkinStyleProvider. This is under 14 instances of FileSystemStyleCache$Entry each with about 6MB of memory
> Heap shows these keys for styles
> Locale Version
> en Mozilla/5.0 (Windows NT 5.1; rv:2.0b7) Gecko/20100101 Firefox/4.0b7
> ko Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US; rv:1.9.1.11) Gecko/20100701 Firefox/3.5.11 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US; rv:1.9.1.11) Gecko/20100701 Firefox/3.5.11 ( .NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/534.10 (KHTML, like Gecko) Chrome/8.0.552.224 Safari/534.10
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/534.10 (KHTML, like Gecko) Chrome/8.0.552.224 Safari/534.10
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.3) Gecko/20090824 Firefox/3.5.3
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.3) Gecko/20090824 Firefox/3.5.3
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.0.18) Gecko/2010020220 Firefox/3.0.18 (.NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.0.18) Gecko/2010020220 Firefox/3.0.18 (.NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.9) Gecko/20100315 Firefox/3.5.9 (.NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.9) Gecko/20100315 Firefox/3.5.9 (.NET CLR 3.5.30729)
> en Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 5.2; Trident/4.0; .NET CLR 1.1.4322; .NET CLR 2.0.50727; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2) Gecko/20100115 Firefox/3.6 (.NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.15) Gecko/20101026 Firefox/3.5.15 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.10) Gecko/20100914 Firefox/3.6.10 (.NET CLR 3.5.30729)
> ko Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; .NET CLR 2.0.50727; .NET CLR 1.1.4322; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; InfoPath.2)
> en Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; .NET CLR 2.0.50727; .NET CLR 1.1.4322; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; InfoPath.2)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.12) Gecko/20101026 Firefox/3.6.12 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.16) Gecko/20101130 Firefox/3.5.16 ( .NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.16) Gecko/20101130 Firefox/3.5.16 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.13) Gecko/20101203 Firefox/3.6.13
> Now beside the amount of memory pinned by single FileSystemStyleCache$Entry the issue is that we apparently create too many different versions and there is no restriction to the size of cache leaving open door for memory leak
> Two questions
> 1. Do we really have different styles for FF on 3.5 or 3.6 or so? If not why do we key by it?
> 2. Given different browsers and locales having cache as plain concurrent hash map is actually a source of leak as over time it can accumulate more and more. Do we have any restriction there? Should the entries by LRU or have exparation
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (TRINIDAD-1985) High live memory usage from
SkinStyleProvider
Posted by "Jeanne Waldman (JIRA)" <de...@myfaces.apache.org>.
[ https://issues.apache.org/jira/browse/TRINIDAD-1985?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12980425#action_12980425 ]
Jeanne Waldman commented on TRINIDAD-1985:
------------------------------------------
When we get a new browser/locale/etc., we get the StyleSheetNodes that match the StyleContext (this is browser,locale,etc).
Then we merge these together to get the final list of StyleNodes, and from this we generate the css file.
This is the code that needs to be re-run when we get a new browser/locale/etc., where the resulting StyleSheetNodes are unique.
In the adf faces skin files, we have many rules that are particular to a browser, even a browser version
@agent gecko and (version: 1.7)
@agent gecko and (max-version: 1.9.0)
@agent gecko and (version: 1.8) {
So for gecko 1.7, we'll generate a css file. For gecko 1.8, we'll generate a different css file, and so on.
Stevan and I think we can try a LRU cache of size 8, and make it configurable so we can run some tests.
(also, consider optimizing the css generation code as mentioned in https://issues.apache.org/jira/browse/TRINIDAD-1675, because if we improve memory with this jira fix, then we'll need to generate the css files more often and we'll want to speed up that code.)
> High live memory usage from SkinStyleProvider
> ---------------------------------------------
>
> Key: TRINIDAD-1985
> URL: https://issues.apache.org/jira/browse/TRINIDAD-1985
> Project: MyFaces Trinidad
> Issue Type: Bug
> Components: Archetype
> Affects Versions: 1.2.12-core
> Reporter: Stevan Malesevic
>
> We were checking a live memory usage in one app and noticed that some 83MB of heap is pinned under SkinStyleProvider. This is under 14 instances of FileSystemStyleCache$Entry each with about 6MB of memory
> Heap shows these keys for styles
> Locale Version
> en Mozilla/5.0 (Windows NT 5.1; rv:2.0b7) Gecko/20100101 Firefox/4.0b7
> ko Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US; rv:1.9.1.11) Gecko/20100701 Firefox/3.5.11 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US; rv:1.9.1.11) Gecko/20100701 Firefox/3.5.11 ( .NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/534.10 (KHTML, like Gecko) Chrome/8.0.552.224 Safari/534.10
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/534.10 (KHTML, like Gecko) Chrome/8.0.552.224 Safari/534.10
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.3) Gecko/20090824 Firefox/3.5.3
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.3) Gecko/20090824 Firefox/3.5.3
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.0.18) Gecko/2010020220 Firefox/3.0.18 (.NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.0.18) Gecko/2010020220 Firefox/3.0.18 (.NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.9) Gecko/20100315 Firefox/3.5.9 (.NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.9) Gecko/20100315 Firefox/3.5.9 (.NET CLR 3.5.30729)
> en Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 5.2; Trident/4.0; .NET CLR 1.1.4322; .NET CLR 2.0.50727; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2) Gecko/20100115 Firefox/3.6 (.NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.15) Gecko/20101026 Firefox/3.5.15 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.10) Gecko/20100914 Firefox/3.6.10 (.NET CLR 3.5.30729)
> ko Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; .NET CLR 2.0.50727; .NET CLR 1.1.4322; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; InfoPath.2)
> en Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; .NET CLR 2.0.50727; .NET CLR 1.1.4322; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; InfoPath.2)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.12) Gecko/20101026 Firefox/3.6.12 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.16) Gecko/20101130 Firefox/3.5.16 ( .NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.16) Gecko/20101130 Firefox/3.5.16 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.13) Gecko/20101203 Firefox/3.6.13
> Now beside the amount of memory pinned by single FileSystemStyleCache$Entry the issue is that we apparently create too many different versions and there is no restriction to the size of cache leaving open door for memory leak
> Two questions
> 1. Do we really have different styles for FF on 3.5 or 3.6 or so? If not why do we key by it?
> 2. Given different browsers and locales having cache as plain concurrent hash map is actually a source of leak as over time it can accumulate more and more. Do we have any restriction there? Should the entries by LRU or have exparation
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Issue Comment Edited: (TRINIDAD-1985) High live memory usage
from SkinStyleProvider
Posted by "Jeanne Waldman (JIRA)" <de...@myfaces.apache.org>.
[ https://issues.apache.org/jira/browse/TRINIDAD-1985?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12977569#action_12977569 ]
Jeanne Waldman edited comment on TRINIDAD-1985 at 1/4/11 9:02 PM:
------------------------------------------------------------------
For me to answer #1, I would need to know what skin you were using, and then I would look at the CSS files for that skin to see if we have different css for FF on 3.5 and 3.6 (like @agent gecko blocks).
I see in the adf faces skins, we do have different styles for gecko 1.9.2 and lower:
/* Firefox versions earlier than 3.6 (Gecko 1.9.2.0) do no support multiple background images */
And we do have slightly different styles for ko in Trinidad's base-desktop.css
We have two types of caches in FileSystemStyleCache. One is am exact match (browser, locale, etc) key, and another is the derivationKey, and that keys off of the styleSheetNodes, so even if you are using different browsers or locales, if they have the same styleSheetNodes, then you will get the same cached value.
What skin were you using?
If we go with a most recently used limit, what do you think a reasonable limit would be?
was (Author: jeanne.waldman@oracle.com):
For me to answer #1, I would need to know what skin you were using, and then I would look at the CSS files for that skin to see if we have different css for FF on 3.5 and 3.6 (like @agent gecko blocks).
What skin were you using?
We have two types of caches in FileSystemStyleCache. One is a derivationKey, and that keys off of the styleSheetNodes, so even if you are using different browsers or locales, if they have the same styleSheetNodes, then you will get the same cached value.
> High live memory usage from SkinStyleProvider
> ---------------------------------------------
>
> Key: TRINIDAD-1985
> URL: https://issues.apache.org/jira/browse/TRINIDAD-1985
> Project: MyFaces Trinidad
> Issue Type: Bug
> Components: Archetype
> Affects Versions: 1.2.12-core
> Reporter: Stevan Malesevic
>
> We were checking a live memory usage in one app and noticed that some 83MB of heap is pinned under SkinStyleProvider. This is under 14 instances of FileSystemStyleCache$Entry each with about 6MB of memory
> Heap shows these keys for styles
> Locale Version
> en Mozilla/5.0 (Windows NT 5.1; rv:2.0b7) Gecko/20100101 Firefox/4.0b7
> ko Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US; rv:1.9.1.11) Gecko/20100701 Firefox/3.5.11 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US; rv:1.9.1.11) Gecko/20100701 Firefox/3.5.11 ( .NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/534.10 (KHTML, like Gecko) Chrome/8.0.552.224 Safari/534.10
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/534.10 (KHTML, like Gecko) Chrome/8.0.552.224 Safari/534.10
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.3) Gecko/20090824 Firefox/3.5.3
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.3) Gecko/20090824 Firefox/3.5.3
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.0.18) Gecko/2010020220 Firefox/3.0.18 (.NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.0.18) Gecko/2010020220 Firefox/3.0.18 (.NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.9) Gecko/20100315 Firefox/3.5.9 (.NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.9) Gecko/20100315 Firefox/3.5.9 (.NET CLR 3.5.30729)
> en Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 5.2; Trident/4.0; .NET CLR 1.1.4322; .NET CLR 2.0.50727; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2) Gecko/20100115 Firefox/3.6 (.NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.15) Gecko/20101026 Firefox/3.5.15 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.10) Gecko/20100914 Firefox/3.6.10 (.NET CLR 3.5.30729)
> ko Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; .NET CLR 2.0.50727; .NET CLR 1.1.4322; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; InfoPath.2)
> en Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; .NET CLR 2.0.50727; .NET CLR 1.1.4322; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; InfoPath.2)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.12) Gecko/20101026 Firefox/3.6.12 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.16) Gecko/20101130 Firefox/3.5.16 ( .NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.16) Gecko/20101130 Firefox/3.5.16 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.13) Gecko/20101203 Firefox/3.6.13
> Now beside the amount of memory pinned by single FileSystemStyleCache$Entry the issue is that we apparently create too many different versions and there is no restriction to the size of cache leaving open door for memory leak
> Two questions
> 1. Do we really have different styles for FF on 3.5 or 3.6 or so? If not why do we key by it?
> 2. Given different browsers and locales having cache as plain concurrent hash map is actually a source of leak as over time it can accumulate more and more. Do we have any restriction there? Should the entries by LRU or have exparation
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] [Commented] (TRINIDAD-1985) High live memory usage from
SkinStyleProvider
Posted by "Jeanne Waldman (JIRA)" <de...@myfaces.apache.org>.
[ https://issues.apache.org/jira/browse/TRINIDAD-1985?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13019442#comment-13019442 ]
Jeanne Waldman commented on TRINIDAD-1985:
------------------------------------------
The LRUCache.patch should be for the related issue TRINIDAD-2026 memory leak in skinstyleprovider. This occurs if one application uses a bunch of different skins.
> High live memory usage from SkinStyleProvider
> ---------------------------------------------
>
> Key: TRINIDAD-1985
> URL: https://issues.apache.org/jira/browse/TRINIDAD-1985
> Project: MyFaces Trinidad
> Issue Type: Bug
> Components: Archetype
> Affects Versions: 1.2.12-core
> Reporter: Stevan Malesevic
> Assignee: Jeanne Waldman
> Attachments: LRUCache.patch
>
>
> We were checking a live memory usage in one app and noticed that some 83MB of heap is pinned under SkinStyleProvider. This is under 14 instances of FileSystemStyleCache$Entry each with about 6MB of memory
> Heap shows these keys for styles
> Locale Version
> en Mozilla/5.0 (Windows NT 5.1; rv:2.0b7) Gecko/20100101 Firefox/4.0b7
> ko Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US; rv:1.9.1.11) Gecko/20100701 Firefox/3.5.11 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US; rv:1.9.1.11) Gecko/20100701 Firefox/3.5.11 ( .NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/534.10 (KHTML, like Gecko) Chrome/8.0.552.224 Safari/534.10
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/534.10 (KHTML, like Gecko) Chrome/8.0.552.224 Safari/534.10
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.3) Gecko/20090824 Firefox/3.5.3
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.3) Gecko/20090824 Firefox/3.5.3
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.0.18) Gecko/2010020220 Firefox/3.0.18 (.NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.0.18) Gecko/2010020220 Firefox/3.0.18 (.NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.9) Gecko/20100315 Firefox/3.5.9 (.NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.9) Gecko/20100315 Firefox/3.5.9 (.NET CLR 3.5.30729)
> en Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 5.2; Trident/4.0; .NET CLR 1.1.4322; .NET CLR 2.0.50727; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2) Gecko/20100115 Firefox/3.6 (.NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.15) Gecko/20101026 Firefox/3.5.15 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.10) Gecko/20100914 Firefox/3.6.10 (.NET CLR 3.5.30729)
> ko Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; .NET CLR 2.0.50727; .NET CLR 1.1.4322; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; InfoPath.2)
> en Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; .NET CLR 2.0.50727; .NET CLR 1.1.4322; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; InfoPath.2)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.12) Gecko/20101026 Firefox/3.6.12 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.16) Gecko/20101130 Firefox/3.5.16 ( .NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.16) Gecko/20101130 Firefox/3.5.16 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.13) Gecko/20101203 Firefox/3.6.13
> Now beside the amount of memory pinned by single FileSystemStyleCache$Entry the issue is that we apparently create too many different versions and there is no restriction to the size of cache leaving open door for memory leak
> Two questions
> 1. Do we really have different styles for FF on 3.5 or 3.6 or so? If not why do we key by it?
> 2. Given different browsers and locales having cache as plain concurrent hash map is actually a source of leak as over time it can accumulate more and more. Do we have any restriction there? Should the entries by LRU or have exparation
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] Commented: (TRINIDAD-1985) High live memory usage from
SkinStyleProvider
Posted by "Jeanne Waldman (JIRA)" <de...@myfaces.apache.org>.
[ https://issues.apache.org/jira/browse/TRINIDAD-1985?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12980516#action_12980516 ]
Jeanne Waldman commented on TRINIDAD-1985:
------------------------------------------
Something to think about --
In the FileSystemStyleCache code, the _cache and _entryCache are ConcurrentMaps. If we use an LRU implementation org.apache.myfaces.trinidadinternal.util.LRUCache extends LinkedHashMap, and wrap this in Collections.synchronizedMap, then we will lose the features of ConcurrentMap. We'll need to synchronize on remove, and from what I've read, we will lose scalability since only one thread can access the Map at once with synchronizedMap.
> High live memory usage from SkinStyleProvider
> ---------------------------------------------
>
> Key: TRINIDAD-1985
> URL: https://issues.apache.org/jira/browse/TRINIDAD-1985
> Project: MyFaces Trinidad
> Issue Type: Bug
> Components: Archetype
> Affects Versions: 1.2.12-core
> Reporter: Stevan Malesevic
>
> We were checking a live memory usage in one app and noticed that some 83MB of heap is pinned under SkinStyleProvider. This is under 14 instances of FileSystemStyleCache$Entry each with about 6MB of memory
> Heap shows these keys for styles
> Locale Version
> en Mozilla/5.0 (Windows NT 5.1; rv:2.0b7) Gecko/20100101 Firefox/4.0b7
> ko Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US; rv:1.9.1.11) Gecko/20100701 Firefox/3.5.11 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US; rv:1.9.1.11) Gecko/20100701 Firefox/3.5.11 ( .NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/534.10 (KHTML, like Gecko) Chrome/8.0.552.224 Safari/534.10
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/534.10 (KHTML, like Gecko) Chrome/8.0.552.224 Safari/534.10
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.3) Gecko/20090824 Firefox/3.5.3
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.3) Gecko/20090824 Firefox/3.5.3
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.0.18) Gecko/2010020220 Firefox/3.0.18 (.NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.0.18) Gecko/2010020220 Firefox/3.0.18 (.NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.9) Gecko/20100315 Firefox/3.5.9 (.NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.9) Gecko/20100315 Firefox/3.5.9 (.NET CLR 3.5.30729)
> en Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 5.2; Trident/4.0; .NET CLR 1.1.4322; .NET CLR 2.0.50727; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2) Gecko/20100115 Firefox/3.6 (.NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.15) Gecko/20101026 Firefox/3.5.15 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.10) Gecko/20100914 Firefox/3.6.10 (.NET CLR 3.5.30729)
> ko Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; .NET CLR 2.0.50727; .NET CLR 1.1.4322; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; InfoPath.2)
> en Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; .NET CLR 2.0.50727; .NET CLR 1.1.4322; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; InfoPath.2)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.12) Gecko/20101026 Firefox/3.6.12 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.16) Gecko/20101130 Firefox/3.5.16 ( .NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.16) Gecko/20101130 Firefox/3.5.16 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.13) Gecko/20101203 Firefox/3.6.13
> Now beside the amount of memory pinned by single FileSystemStyleCache$Entry the issue is that we apparently create too many different versions and there is no restriction to the size of cache leaving open door for memory leak
> Two questions
> 1. Do we really have different styles for FF on 3.5 or 3.6 or so? If not why do we key by it?
> 2. Given different browsers and locales having cache as plain concurrent hash map is actually a source of leak as over time it can accumulate more and more. Do we have any restriction there? Should the entries by LRU or have exparation
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (TRINIDAD-1985) High live memory usage from
SkinStyleProvider
Posted by "Jeanne Waldman (JIRA)" <de...@myfaces.apache.org>.
[ https://issues.apache.org/jira/browse/TRINIDAD-1985?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12977592#action_12977592 ]
Jeanne Waldman commented on TRINIDAD-1985:
------------------------------------------
Filed this related JIRA issue: TRINIDAD-1994 remove unused @locale selectors to improve skinning performance
> High live memory usage from SkinStyleProvider
> ---------------------------------------------
>
> Key: TRINIDAD-1985
> URL: https://issues.apache.org/jira/browse/TRINIDAD-1985
> Project: MyFaces Trinidad
> Issue Type: Bug
> Components: Archetype
> Affects Versions: 1.2.12-core
> Reporter: Stevan Malesevic
>
> We were checking a live memory usage in one app and noticed that some 83MB of heap is pinned under SkinStyleProvider. This is under 14 instances of FileSystemStyleCache$Entry each with about 6MB of memory
> Heap shows these keys for styles
> Locale Version
> en Mozilla/5.0 (Windows NT 5.1; rv:2.0b7) Gecko/20100101 Firefox/4.0b7
> ko Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US; rv:1.9.1.11) Gecko/20100701 Firefox/3.5.11 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US; rv:1.9.1.11) Gecko/20100701 Firefox/3.5.11 ( .NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/534.10 (KHTML, like Gecko) Chrome/8.0.552.224 Safari/534.10
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/534.10 (KHTML, like Gecko) Chrome/8.0.552.224 Safari/534.10
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.3) Gecko/20090824 Firefox/3.5.3
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.3) Gecko/20090824 Firefox/3.5.3
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.0.18) Gecko/2010020220 Firefox/3.0.18 (.NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.0.18) Gecko/2010020220 Firefox/3.0.18 (.NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.9) Gecko/20100315 Firefox/3.5.9 (.NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.9) Gecko/20100315 Firefox/3.5.9 (.NET CLR 3.5.30729)
> en Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 5.2; Trident/4.0; .NET CLR 1.1.4322; .NET CLR 2.0.50727; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2) Gecko/20100115 Firefox/3.6 (.NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.15) Gecko/20101026 Firefox/3.5.15 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.10) Gecko/20100914 Firefox/3.6.10 (.NET CLR 3.5.30729)
> ko Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; .NET CLR 2.0.50727; .NET CLR 1.1.4322; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; InfoPath.2)
> en Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; .NET CLR 2.0.50727; .NET CLR 1.1.4322; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; InfoPath.2)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.12) Gecko/20101026 Firefox/3.6.12 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.16) Gecko/20101130 Firefox/3.5.16 ( .NET CLR 3.5.30729)
> ko Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.16) Gecko/20101130 Firefox/3.5.16 ( .NET CLR 3.5.30729)
> en Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.13) Gecko/20101203 Firefox/3.6.13
> Now beside the amount of memory pinned by single FileSystemStyleCache$Entry the issue is that we apparently create too many different versions and there is no restriction to the size of cache leaving open door for memory leak
> Two questions
> 1. Do we really have different styles for FF on 3.5 or 3.6 or so? If not why do we key by it?
> 2. Given different browsers and locales having cache as plain concurrent hash map is actually a source of leak as over time it can accumulate more and more. Do we have any restriction there? Should the entries by LRU or have exparation
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.