1.4.2 - Test 2 (Was: Re: Re: [JDEV] reproducible crash)

Mon Feb 4 13:58:48 CST 2002

There are very few users, possibly 2 or 3 log in over the period of a
day, with one user always connected.

This snapshot was taken at random with only one user logged in.

I've just taken another random snapshot and got;

 19:53:53 up 103 days, 23:29,  4 users,  load average: 0.45, 0.51, 0.42
227 processes: 213 sleeping, 5 running, 8 zombie, 1 stopped
CPU states:  46.2% user,   6.2% system,   2.9% nice,  44.8% idle
Mem:    647252K total,   644776K used,     2476K free,    13156K buffers
Swap:   996020K total,   449104K used,   546916K free,   129416K cached

  PID USER     PRI  NI  SIZE  RSS SHARE STAT %CPU %MEM   TIME COMMAND
31864 alsutton  15   0  1888 1888   636 R    20.3  0.2  68:41 jabberd

and I've experienced load levels >10% of CPU on almost every check I've
done (I disabled all transfers because the machine owners were
complaining it was maxing out one of the CPUs).

Al.

On Mon, 2002-02-04 at 17:58, Michael F. March wrote:
> Approximately how many concurrent users do you have logged on
> when the load is that high?
> 
> > Does CVS also include fixes for jabberd eating CPU like a hog?
> >
> > I've been running 1.4.2-test2 without any transports and after less than
> 48hrs this is my top output;
> >
> >  17:28:31 up 103 days, 21:04,  7 users,  load average: 0.99, 1.18, 1.32
> > 245 processes: 237 sleeping, 2 running, 5 zombie, 1 stopped
> > CPU states:  37.3% user,   8.1% system,   0.0% nice,  54.6% idle
> > Mem:    647252K total,   643404K used,     3848K free,    17064K buffers
> > Swap:   996020K total,   449696K used,   546324K free,   124376K cached
> >
> >   PID USER     PRI  NI  SIZE  RSS SHARE STAT %CPU %MEM   TIME COMMAND
> > 31864 alsutton  14   0  1748 1748   620 S    26.3  0.2  46:58 jabberd
> >
> >
> > :(.
> >
> > Al.
> >
> > >> Original Message from "Thomas Muldowney" <temas at box5.net> on Mon, 4 Feb
> 2002 11:09:52 -0600 <<
> >
> > Definately upgrade your server.  There have been a ton of changes in CVS
> > since that time.
> >
> > --temas
> >
> >
> > ----- Original Message -----
> > From: tom_waters at mac.com
> > To: jdev at jabber.org
> > Sent: Saturday, February 02, 2002 9:43 PM
> > Subject: [JDEV] reproducible crash
> >
> >
> > My client wants to be able to create new user accounts after the user has
> > logged in...
> > I'm not sure if this is supposed to be allowed, but when I have the
> > following exchange with the server (build from CVS on 1/8/02)
> >
> > <?xml version='1.0'?><stream:stream
> > xmlns:stream='http://etherx.jabber.org/streams' id='3C5CAF04'
> > xmlns='jabber:client' from='myservername'>
> >
> > <iq type="set" id="1001"><query
> >
> xmlns="jabber:iq:auth"><username>waters</username><password>jabber</password
> > ><resource>InstantPictures</resource></query></iq>
> >
> > <iq type='result' id='1001'/>
> >
> > <iq type="get" id="1002"><query xmlns="jabber:iq:register"/></iq>
> >
> > I get a segfault with the following jabberd -D output
> >
> > Sat Feb 2 19:39:54 2002 mio.c:608 mio while loop, working
> > Sat Feb 2 19:39:54 2002 mio.c:746 MIO read from socket 10: <iq type="get"
> > id="1002"><query xmlns="jabber:iq:register"/></iq>
> > Sat Feb 2 19:39:54 2002 c2s [client.c:266] pthsock_client_read called
> with:
> > m:1A21D0 flag:3 arg:1A22F0
> > Sat Feb 2 19:39:54 2002 deliver.c:460 DELIVER 4:myservername <route
> > to='waters at myservername/1AAB90' from='10 at c2s/1A21D0'><iq type='get'
> > id='1002'><query xmlns='jabber:iq:register'/></iq></route>
> > Sat Feb 2 19:39:54 2002 deliver.c:664 delivering to instance 'sessions'
> > Sat Feb 2 19:39:54 2002 deliver.c:83 (17CB50)incoming packet <route
> > to='waters at myservername/1AAB90' from='10 at c2s/1A21D0'><iq type='get'
> > id='1002'><query xmlns='jabber:iq:register'/></iq></route>
> > Sat Feb 2 19:39:54 2002 users.c:132 js_user(waters at myservername,205000)
> > Sat Feb 2 19:39:54 2002 mtq 1A5620 entering from pth
> > Sat Feb 2 19:39:54 2002 mtq 1A5620 queue call 1A87D0
> > Sat Feb 2 19:39:54 2002 sessions.c:235 THREAD:SESSION:FROM received a
> > packet!
> > Sat Feb 2 19:39:54 2002 modules.c:124 mapi_call 1
> > Sat Feb 2 19:39:54 2002 modules.c:147 MAPI 1AAE40
> > Sat Feb 2 19:39:54 2002 modules.c:147 MAPI 1AAE50
> > Sat Feb 2 19:39:54 2002 modules.c:147 MAPI 1AAE80
> > Sat Feb 2 19:39:54 2002 modules.c:147 MAPI 1AAE90
> > Sat Feb 2 19:39:54 2002 modules.c:147 MAPI 1AAEA0
> > Sat Feb 2 19:39:54 2002 mod_filter.c:791
> > packet out from mod_filter
> >
> > Sat Feb 2 19:39:54 2002 modules.c:147 MAPI 1AAEB0
> > Sat Feb 2 19:39:54 2002 modules.c:147 MAPI 1AAF90
> > Sat Feb 2 19:39:54 2002 modules.c:147 MAPI 1AAFA0
> > Sat Feb 2 19:39:54 2002 modules.c:147 MAPI 1AAFD0
> > Sat Feb 2 19:39:54 2002 modules.c:164 mapi_call returning unhandled
> > Sat Feb 2 19:39:54 2002 dnsrv.c:144 dnsrv: Read error on coprocess(1): 1
> > Operation not permitted
> > Sat Feb 2 19:39:54 2002 dnsrv.c:158 DNSRV CHILD: out of loop.. exiting
> > normal
> > Segmentation fault
> >
> > _______________________________________________
> > jdev mailing list
> > jdev at jabber.org
> > http://mailman.jabber.org/listinfo/jdev
> >
> > --
> > Al Sutton
> > al at alsutton.com
> > _______________________________________________
> > jdev mailing list
> > jdev at jabber.org
> > http://mailman.jabber.org/listinfo/jdev
> 
> _______________________________________________
> jdev mailing list
> jdev at jabber.org
> http://mailman.jabber.org/listinfo/jdev