diff options
Diffstat (limited to 'zarb-ml/mageia-sysadm/2011-March/003027.html')
| -rw-r--r-- | zarb-ml/mageia-sysadm/2011-March/003027.html | 140 |
1 files changed, 140 insertions, 0 deletions
diff --git a/zarb-ml/mageia-sysadm/2011-March/003027.html b/zarb-ml/mageia-sysadm/2011-March/003027.html new file mode 100644 index 000000000..622fd5796 --- /dev/null +++ b/zarb-ml/mageia-sysadm/2011-March/003027.html @@ -0,0 +1,140 @@ +<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN"> +<HTML> + <HEAD> + <TITLE> [Mageia-sysadm] Alamut memory problem + </TITLE> + <LINK REL="Index" HREF="index.html" > + <LINK REL="made" HREF="mailto:mageia-sysadm%40mageia.org?Subject=Re%3A%20%5BMageia-sysadm%5D%20Alamut%20memory%20problem&In-Reply-To=%3C1299582078.8638.13.camel%40akroma.ephaone.org%3E"> + <META NAME="robots" CONTENT="index,nofollow"> + <META http-equiv="Content-Type" content="text/html; charset=us-ascii"> + <LINK REL="Previous" HREF="003021.html"> + <LINK REL="Next" HREF="003022.html"> + </HEAD> + <BODY BGCOLOR="#ffffff"> + <H1>[Mageia-sysadm] Alamut memory problem</H1> + <B>Michael Scherer</B> + <A HREF="mailto:mageia-sysadm%40mageia.org?Subject=Re%3A%20%5BMageia-sysadm%5D%20Alamut%20memory%20problem&In-Reply-To=%3C1299582078.8638.13.camel%40akroma.ephaone.org%3E" + TITLE="[Mageia-sysadm] Alamut memory problem">misc at zarb.org + </A><BR> + <I>Tue Mar 8 12:01:18 CET 2011</I> + <P><UL> + <LI>Previous message: <A HREF="003021.html">[Mageia-sysadm] Alamut memory problem +</A></li> + <LI>Next message: <A HREF="003022.html">[Mageia-sysadm] Proposal for maintainers database API +</A></li> + <LI> <B>Messages sorted by:</B> + <a href="date.html#3027">[ date ]</a> + <a href="thread.html#3027">[ thread ]</a> + <a href="subject.html#3027">[ subject ]</a> + <a href="author.html#3027">[ author ]</a> + </LI> + </UL> + <HR> +<!--beginarticle--> +<PRE>Le dimanche 06 mars 2011 à 12:57 +0100, Michael Scherer a écrit : +><i> Hi, +</I>><i> +</I>><i> it seems that alamut memory usage is growing since a few days ( in +</I>><i> fact, we have seen it now that we monitor it thanks to xymon +</I>><i> installation, but it should be since a much longer time ). +</I>><i> +</I>><i> After some hours of debugging with dmorgan, I narrowed the problem down +</I>><i> to viewvc, after careful looking to logs and with the help of watch -d +</I>><i> & +</I>><i> ps .Each time we see a line IOError in error.log, that the sign +</I>><i> there is a memory leaked ( as we can see on graphs if we look carefully +</I>><i> ). I +</I>><i> suspect that the problem is on viewvc side, as explained on : +</I>><i> +</I>><i> <A HREF="http://viewvc.tigris.org/ds/viewMessage.do?dsForumId=4254&dsMessageId=2364399">http://viewvc.tigris.org/ds/viewMessage.do?dsForumId=4254&dsMessageId=2364399</A> +</I>><i> +</I>><i> Before, we did allowed wsgi web software to register custom signals +</I>><i> handler ( +</I>><i> <A HREF="http://svnweb.mageia.org/adm/puppet/modules/apache/templates/mod_wsgi.conf?r1=908&r2=1051">http://svnweb.mageia.org/adm/puppet/modules/apache/templates/mod_wsgi.conf?r1=908&r2=1051</A> +</I>><i> ), but this caused apache to not restart, and so it was disabled again( +</I>><i> <A HREF="http://svnweb.mageia.org/adm/puppet/modules/apache/templates/mod_wsgi.conf?r1=1051&r2=1195">http://svnweb.mageia.org/adm/puppet/modules/apache/templates/mod_wsgi.conf?r1=1051&r2=1195</A> +</I>><i> ). +</I>><i> +</I>><i> So basically, when we see this : +</I>><i> +</I>><i> [Tue Mar 01 07:31:43 2011] [error] [client 74.125.16.66] Traceback +</I>><i> (most recent call last): +</I>><i> [Tue Mar 01 07:31:43 2011] [error] [client 74.125.16.66] File +</I>><i> "/usr/share/viewvc/bin/wsgi/viewvc.wsgi", line 40, in application +</I>><i> [Tue Mar 01 07:31:43 2011] [error] [client 74.125.16.66] +</I>><i> viewvc.main(server, cfg) +</I>><i> [Tue Mar 01 07:31:43 2011] [error] [client 74.125.16.66] File +</I>><i> "/usr/share/viewvc/lib/viewvc.py", line 4415, in main +</I>><i> [Tue Mar 01 07:31:43 2011] [error] [client 74.125.16.66] +</I>><i> view_error(server, cfg) +</I>><i> [Tue Mar 01 07:31:43 2011] [error] [client 74.125.16.66] File +</I>><i> "/usr/share/viewvc/lib/viewvc.py", line 4403, in view_error +</I>><i> [Tue Mar 01 07:31:43 2011] [error] [client 74.125.16.66] +</I>><i> debug.PrintException(server, exc_dict) +</I>><i> [Tue Mar 01 07:31:43 2011] [error] [client 74.125.16.66] File +</I>><i> "/usr/share/viewvc/lib/debug.py", line 81, in PrintException +</I>><i> [Tue Mar 01 07:31:43 2011] [error] [client 74.125.16.66] +</I>><i> server.write("<h3>An Exception Has Occurred</h3>\\n") +</I>><i> [Tue Mar 01 07:31:43 2011] [error] [client 74.125.16.66] File +</I>><i> "/usr/share/viewvc/lib/sapi.py", line 252, in write +</I>><i> [Tue Mar 01 07:31:43 2011] [error] [client 74.125.16.66] +</I>><i> self._wsgi_write(s) +</I>><i> [Tue Mar 01 07:31:43 2011] [error] [client 74.125.16.66] IOError: +</I>><i> client connection closed +</I>><i> +</I>><i> there is a problem. +</I>><i> +</I>><i> At first, I thought it was some users with crappy internet connexion, +</I>><i> but this example is googlebot. +</I>><i> +</I>><i> So far, we have a few options : +</I>><i> - going back on cgi would mask the errors, to the cost of speed +</I>><i> - asking to apache to recycle more often his childs would also mask the +</I>><i> error +</I>><i> - reallowing mod_wsgi to add a signal handler would likely do it too +</I>><i> - fixing the leak on viewvc side should IMHO better but maybe too +</I>><i> complex +</I>><i> +</I>><i> So I propose the following : +</I>><i> +</I>><i> - use python-flup to make viewvc run a external fastcgi process, so it +</I>><i> can register signal handler as he see fit, and do not interfer with +</I>><i> apache. This also bring consistancy ( since we use fastcgi for almost +</I>><i> everything, except bugzilla and django applications ), and would allow +</I>><i> a finer grained control ( as we can see each process taking memory and +</I>><i> plan ressources based on this ). +</I>><i> +</I>><i> WDYT ? +</I> +Since no one commented, I went ahead this morning. So far, it still work +fine, we will see if this help. + +-- +Michael Scherer + +</PRE> + + + + + +<!--endarticle--> + <HR> + <P><UL> + <!--threads--> + <LI>Previous message: <A HREF="003021.html">[Mageia-sysadm] Alamut memory problem +</A></li> + <LI>Next message: <A HREF="003022.html">[Mageia-sysadm] Proposal for maintainers database API +</A></li> + <LI> <B>Messages sorted by:</B> + <a href="date.html#3027">[ date ]</a> + <a href="thread.html#3027">[ thread ]</a> + <a href="subject.html#3027">[ subject ]</a> + <a href="author.html#3027">[ author ]</a> + </LI> + </UL> + +<hr> +<a href="https://www.mageia.org/mailman/listinfo/mageia-sysadm">More information about the Mageia-sysadm +mailing list</a><br> +</body></html> |