You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@activemq.apache.org by Fredrik Jonson <fr...@jonson.org> on 2010/09/03 07:47:29 UTC
Add a progress indicator to the file cursor recovery phase
Hello,
I saw AMQ-2891 passing by in the mail flood, and it reminded me of a feature
I'd like to see in activemq. I started to comment in the issue, but decided to
post to the dev list instead. https://issues.apache.org/activemq/browse/AMQ-2891
This is not a comment on the actual issue, but a possible usability workaround.
As a administrator of activemq in production environments my experience is that
it is quite unnerving to watch the deep silence of activemq when restarting
after a fatal failure - especially when you have a huge message log of important
data. I know the cursor recovery is busy somewhere during startup but how can I
tell that it has started, is ok, and how much time it will take?
I think it makes sense to add some sort of progress indicator, as logging
output on the info level, during the recovery phase. Say by logging every 10 000
message recovered, and if that information is available also the total amount
that needs to be recovered.
Does that make sense?
--
Fredrik Jonson
Re: Add a progress indicator to the file cursor recovery phase
Posted by Fredrik Jonson <fr...@jonson.org>.
Thank you, Gary!
That's just awesome.
--
Fredrik Jonson
Re: Add a progress indicator to the file cursor recovery phase
Posted by Gary Tully <ga...@gmail.com>.
done: see https://issues.apache.org/activemq/browse/AMQ-2891?focusedCommentId=61659&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_61659
On 3 September 2010 09:31, Gary Tully <ga...@gmail.com> wrote:
> yea, that makes perfect sense. Think logging every X messages will be
> easy, getting some sort of "precent complete output" will be a little
> harder but an approximation should be possible.
>
> On 3 September 2010 06:47, Fredrik Jonson <fr...@jonson.org> wrote:
>> Hello,
>>
>> I saw AMQ-2891 passing by in the mail flood, and it reminded me of a feature
>> I'd like to see in activemq. I started to comment in the issue, but decided to
>> post to the dev list instead. https://issues.apache.org/activemq/browse/AMQ-2891
>>
>> This is not a comment on the actual issue, but a possible usability workaround.
>>
>> As a administrator of activemq in production environments my experience is that
>> it is quite unnerving to watch the deep silence of activemq when restarting
>> after a fatal failure - especially when you have a huge message log of important
>> data. I know the cursor recovery is busy somewhere during startup but how can I
>> tell that it has started, is ok, and how much time it will take?
>>
>> I think it makes sense to add some sort of progress indicator, as logging
>> output on the info level, during the recovery phase. Say by logging every 10 000
>> message recovered, and if that information is available also the total amount
>> that needs to be recovered.
>>
>> Does that make sense?
>>
>> --
>> Fredrik Jonson
>>
>>
>
>
>
> --
> http://blog.garytully.com
>
> Open Source Integration
> http://fusesource.com
>
--
http://blog.garytully.com
Open Source Integration
http://fusesource.com
Re: Add a progress indicator to the file cursor recovery phase
Posted by Gary Tully <ga...@gmail.com>.
yea, that makes perfect sense. Think logging every X messages will be
easy, getting some sort of "precent complete output" will be a little
harder but an approximation should be possible.
On 3 September 2010 06:47, Fredrik Jonson <fr...@jonson.org> wrote:
> Hello,
>
> I saw AMQ-2891 passing by in the mail flood, and it reminded me of a feature
> I'd like to see in activemq. I started to comment in the issue, but decided to
> post to the dev list instead. https://issues.apache.org/activemq/browse/AMQ-2891
>
> This is not a comment on the actual issue, but a possible usability workaround.
>
> As a administrator of activemq in production environments my experience is that
> it is quite unnerving to watch the deep silence of activemq when restarting
> after a fatal failure - especially when you have a huge message log of important
> data. I know the cursor recovery is busy somewhere during startup but how can I
> tell that it has started, is ok, and how much time it will take?
>
> I think it makes sense to add some sort of progress indicator, as logging
> output on the info level, during the recovery phase. Say by logging every 10 000
> message recovered, and if that information is available also the total amount
> that needs to be recovered.
>
> Does that make sense?
>
> --
> Fredrik Jonson
>
>
--
http://blog.garytully.com
Open Source Integration
http://fusesource.com