You are viewing a plain text version of this content. The canonical link for it is here.
Posted to bugs@httpd.apache.org by bu...@apache.org on 2017/01/05 08:55:04 UTC
[Bug 60553] New: After segmentation fault balancer do not work
properly
https://bz.apache.org/bugzilla/show_bug.cgi?id=60553
Bug ID: 60553
Summary: After segmentation fault balancer do not work properly
Product: Apache httpd-2
Version: 2.4.25
Hardware: PC
OS: Linux
Status: NEW
Severity: normal
Priority: P2
Component: mod_proxy_hcheck
Assignee: bugs@httpd.apache.org
Reporter: egandarias@gfi.es
Target Milestone: ---
Created attachment 34593
--> https://bz.apache.org/bugzilla/attachment.cgi?id=34593&action=edit
Three files with coredump and error log
S.O Version:
Red Hat Enterprise Linux Server release 7.2 (Maipo)
uname -a
Linux ejld1132.ejgvdns 3.10.0-327.13.1.el7.x86_64 #1 SMP Mon Feb 29 13:22:02
EST 2016 x86_64 x86_64 x86_64 GNU/Linux
Apache httpd version:
/opt/apache2/bin/httpd -v
Server version: Apache/2.4.25 (Unix)
Server built: Jan 4 2017 09:34:36
/opt/apache2/bin/httpd -l
Compiled in modules:
core.c
mod_so.c
http_core.c
worker.c
-----------------------------------------------------------
Problem:
When we run apache httpd with a balancer configured
<Proxy balancer://cluster_namenode>
BalancerMember http://ejld1132.ejgvdns:8170 hcmethod=GET hcinterval=5
hcpasses=1 hcfails=1 hcuri=/webhdfs/v1/data?op=GETFILESTATUS
BalancerMember http://ejld1131.ejgvdns:8170 hcmethod=GET hcinterval=5
hcpasses=1 hcfails=1 hcuri=/webhdfs/v1/data?op=GETFILESTATUS
</Proxy>
We are using this balancer to deal with the hdfs namenode high availability.
Usually ejld1131 is the active one, and the other is in standby mode. So
requests to ejld1132 responds with 403 code and is marked as DISABLED. Ejld1131
responds with 200 code.
it starts working properly. We have make tests with jmeter with good results
but sometimes "Segmentation Fault" alert appears on error log.
Anyway it continues working but not everytime. After segmentation fault
sometimes mod_proxy_hcheck doesn't find defined workers properly and requests
don't work. After another "Segmentation Fault" workers are restarted again and
it starts working well.
Fragments of error log with this problem located are attached.
Thanks
--
You are receiving this mail because:
You are the assignee for the bug.
---------------------------------------------------------------------
To unsubscribe, e-mail: bugs-unsubscribe@httpd.apache.org
For additional commands, e-mail: bugs-help@httpd.apache.org
[Bug 60553] After segmentation fault balancer do not work properly
Posted by bu...@apache.org.
https://bz.apache.org/bugzilla/show_bug.cgi?id=60553
Yann Ylavic <yl...@gmail.com> changed:
What |Removed |Added
----------------------------------------------------------------------------
Resolution|--- |DUPLICATE
Status|NEEDINFO |RESOLVED
--- Comment #7 from Yann Ylavic <yl...@gmail.com> ---
Both crashes likely related, joining the two PRs, and following up there.
*** This bug has been marked as a duplicate of bug 60071 ***
--
You are receiving this mail because:
You are the assignee for the bug.
---------------------------------------------------------------------
To unsubscribe, e-mail: bugs-unsubscribe@httpd.apache.org
For additional commands, e-mail: bugs-help@httpd.apache.org
[Bug 60553] After segmentation fault balancer do not work properly
Posted by bu...@apache.org.
https://bz.apache.org/bugzilla/show_bug.cgi?id=60553
--- Comment #6 from Yann Ylavic <yl...@gmail.com> ---
Could you try the patch in bug 60071 which is possibly related?
--
You are receiving this mail because:
You are the assignee for the bug.
---------------------------------------------------------------------
To unsubscribe, e-mail: bugs-unsubscribe@httpd.apache.org
For additional commands, e-mail: bugs-help@httpd.apache.org
[Bug 60553] After segmentation fault balancer do not work properly
Posted by bu...@apache.org.
https://bz.apache.org/bugzilla/show_bug.cgi?id=60553
--- Comment #3 from Endika <eg...@gfi.es> ---
Hello,
For testing purpose there is no problem, because rigth now we are running this
build in dev environment, but we should have an stable version for next
environments.
About the second question...that's the problem we have found...
[Wed Jan 04 07:25:22.072667 2017] [proxy_hcheck:debug] [pid 18267:tid
140685705389824] mod_proxy_hcheck.c(423): AH03248: Creating hc worker 7ebd00
for http://ejld1131.ejgvdns:8170
[Wed Jan 04 07:25:22.072675 2017] [proxy_hcheck:debug] [pid 18267:tid
140685696997120] mod_proxy_hcheck.c(423): AH03248: Creating hc worker 7eb720
for http://ejld1132.ejgvdns:8170
It should take this name: 7ebd00
but takes the weird name: A\xaa\xadz\xc9"\xc3Njld1131
Other times takes the same name twice (takes 16f36a0 but not 16f3c80)
[Wed Jan 04 12:17:02.455159 2017] [proxy_hcheck:debug] [pid 24207:tid
139728036820736] mod_proxy_hcheck.c(423): AH03248: Creating hc worker 16f36a0
for http://ejld1132.ejgvdns:8170
[Wed Jan 04 12:17:02.455162 2017] [proxy_hcheck:debug] [pid 24207:tid
139728254834432] mod_proxy_hcheck.c(423): AH03248: Creating hc worker 16f3c80
for http://ejld1131.ejgvdns:8170
[Wed Jan 04 12:17:02.455188 2017] [proxy:debug] [pid 24207:tid 139728036820736]
proxy_util.c(1779): AH00925: initializing worker 16f36a0 shared
[Wed Jan 04 12:17:02.455188 2017] [proxy:debug] [pid 24207:tid 139728254834432]
proxy_util.c(1779): AH00925: initializing worker 16f36a0 shared
[Wed Jan 04 12:17:02.455195 2017] [proxy:debug] [pid 24207:tid 139728036820736]
proxy_util.c(1821): AH00927: initializing worker 16f36a0 local
[Wed Jan 04 12:17:02.455200 2017] [proxy:debug] [pid 24207:tid 139728254834432]
proxy_util.c(1821): AH00927: initializing worker 16f36a0 local
The example in info1.txt is when a Segmentation fault occurs but it continues
working correctly
Thank
--
You are receiving this mail because:
You are the assignee for the bug.
---------------------------------------------------------------------
To unsubscribe, e-mail: bugs-unsubscribe@httpd.apache.org
For additional commands, e-mail: bugs-help@httpd.apache.org
[Bug 60553] After segmentation fault balancer do not work properly
Posted by bu...@apache.org.
https://bz.apache.org/bugzilla/show_bug.cgi?id=60553
--- Comment #2 from Jim Jagielski <ji...@apache.org> ---
I cannot see, based on the snippet below, where httpd is creating these weirdly
named workers:
[Wed Jan 04 07:25:22.072715 2017] [proxy:debug] [pid 18267:tid 140685705389824]
proxy_util.c(1779): AH00925: initializing worker A\xaa\xadz\xc9"\xc3Njld1131.
shared
[Wed Jan 04 07:25:22.072717 2017] [proxy:debug] [pid 18267:tid 140685696997120]
proxy_util.c(1779): AH00925: initializing worker 7eb720 shared
[Wed Jan 04 07:25:22.072722 2017] [proxy:debug] [pid 18267:tid 140685705389824]
proxy_util.c(1821): AH00927: initializing worker A\xaa\xadz\xc9"\xc3Njld1131.
local
[Wed Jan 04 07:25:22.072725 2017] [proxy:debug] [pid 18267:tid 140685696997120]
proxy_util.c(1821): AH00927: initializing worker 7eb720 local
--
You are receiving this mail because:
You are the assignee for the bug.
---------------------------------------------------------------------
To unsubscribe, e-mail: bugs-unsubscribe@httpd.apache.org
For additional commands, e-mail: bugs-help@httpd.apache.org
[Bug 60553] After segmentation fault balancer do not work properly
Posted by bu...@apache.org.
https://bz.apache.org/bugzilla/show_bug.cgi?id=60553
Endika <eg...@gfi.es> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |egandarias@gfi.es
--
You are receiving this mail because:
You are the assignee for the bug.
---------------------------------------------------------------------
To unsubscribe, e-mail: bugs-unsubscribe@httpd.apache.org
For additional commands, e-mail: bugs-help@httpd.apache.org
[Bug 60553] After segmentation fault balancer do not work properly
Posted by bu...@apache.org.
https://bz.apache.org/bugzilla/show_bug.cgi?id=60553
--- Comment #5 from Endika <eg...@gfi.es> ---
Hi,
What do you mean about verbose level? Our Apache LogLevel is already in debug
mode
#
# LogLevel: Control the number of messages logged to the error_log.
# Possible values include: debug, info, notice, warn, error, crit,
# alert, emerg.
#
LogLevel debug
Thanks
--
You are receiving this mail because:
You are the assignee for the bug.
---------------------------------------------------------------------
To unsubscribe, e-mail: bugs-unsubscribe@httpd.apache.org
For additional commands, e-mail: bugs-help@httpd.apache.org
[Bug 60553] After segmentation fault balancer do not work properly
Posted by bu...@apache.org.
https://bz.apache.org/bugzilla/show_bug.cgi?id=60553
--- Comment #1 from Jim Jagielski <ji...@apache.org> ---
Is there anyway you could run with a build that lacks optimization?
--
You are receiving this mail because:
You are the assignee for the bug.
---------------------------------------------------------------------
To unsubscribe, e-mail: bugs-unsubscribe@httpd.apache.org
For additional commands, e-mail: bugs-help@httpd.apache.org
[Bug 60553] After segmentation fault balancer do not work properly
Posted by bu...@apache.org.
https://bz.apache.org/bugzilla/show_bug.cgi?id=60553
Jim Jagielski <ji...@apache.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |NEEDINFO
--- Comment #4 from Jim Jagielski <ji...@apache.org> ---
Try running under a verbose loglevel. Thx!
--
You are receiving this mail because:
You are the assignee for the bug.
---------------------------------------------------------------------
To unsubscribe, e-mail: bugs-unsubscribe@httpd.apache.org
For additional commands, e-mail: bugs-help@httpd.apache.org