Quin Guin wrote:
Hi,
 More information on 3.1.0.8 squid restarting on it own issue that I am having.

I had a 111 of the errors for "Internal error: lost DNS error" before I 
shutdown 3.1 and went back to 2.7. I think this might be related to the 3.1 resets but 
what can I do to collect more information to help resolve this issue or if its a known 
issue would someone please point me in the right direction. The system has 8GB RAM and is 
not swapping at all I have 50% of free disk on each JBOD that I am using as a cache 
store. I am only logging to cache.log so I do not hae any log file size issues.

2009/06/01 15:05:10| Internal error: lost DNS error info

2009/06/01 15:19:55| errorpage.cc(293) errorTryLoadText: '/usr/local/squid-3.1.0
.8/share/errors/es-ve/ERR_DNS_FAIL': (2) No such file or directory
2009/06/01 15:19:55| WARNING: Error Pages Missing Language: es-ve
2009/06/01 15:24:33| assertion failed: MemBuf.cc:391: "new_cap > (size_t) capaci
ty"
2009/06/01 15:24:36| Starting Squid Cache version 3.1.0.8-20090531 for x86_64-un
known-linux-gnu...
2009/06/01 15:24:36| Process ID 16274
2009/06/01 15:24:36| With 16384 file descriptors available
2009/06/01 15:24:36| Initializing IP Cache...
2009/06/01 15:24:36| DNS Socket created at 0.0.0.0, FD 7
2009/06/01 15:24:36| Adding nameserver 172.22.0.33 from /etc/resolv..conf
2009/06/01 15:24:36| Adding nameserver 172.22.0.34 from /etc/resolv..conf
2009/06/01 15:24:37| Unlinkd pipe opened on FD 11
2009/06/01 15:24:37| Store logging disabled
2009/06/01 15:24:37| Swap maxSize 122880000 + 262144 KB, estimated 9472472 
objects
2009/06/01 15:24:37| Target number of buckets: 473623
2009/06/01 15:24:37| Using 524288 Store buckets
2009/06/01 15:24:37| Max Mem  size: 262144 KB
2009/06/01 15:24:37| Max Swap size: 122880000 KB
2009/06/01 15:24:37| Version 1 of swap file without LFS support detected... 2009/06/01 15:24:37| Rebuilding storage in /cache1/cache3108 (DIRTY) 2009/06/01 15:24:37| Version 1 of swap file without LFS support detected... 2009/06/01 15:24:37| Rebuilding storage in /cache2/cache3108 (DIRTY) 2009/06/01 15:24:37| Version 1 of swap file without LFS support detected... 2009/06/01 15:24:37| Rebuilding storage in /cache3/cache3108 (DIRTY)
2009/06/01 15:24:37| Using Least Load store dir selection
2009/06/01 15:24:37| Set Current Directory to /usr/local/squid-3.1.0.8/var/cache
2009/06/01 15:24:37| Loaded Icons.
2009/06/01 15:24:37| Accepting  HTTP connections at 0.0.0.0:3128, FD 18.
2009/06/01 15:24:37| HTCP Disabled.
2009/06/01 15:24:37| Squid modules loaded: 0
2009/06/01 15:24:37| Ready to serve requests.
2009/06/01 15:24:37| Store rebuilding is 1.10% complete
2009/06/01 15:24:39| Done reading /cache3/cache3108 swaplog (370238 entries)
2009/06/01 15:24:39| Done reading /cache1/cache3108 swaplog (372615 entries)
2009/06/01 15:24:39| Done reading /cache2/cache3108 swaplog (373734 entries)
2009/06/01 15:24:39| Finished rebuilding storage from disk.
2009/06/01 15:24:39|    922730 Entries scanned
2009/06/01 15:24:39|         0 Invalid entries.
2009/06/01 15:24:39|         0 With invalid flags.
2009/06/01 15:24:39|    761969 Objects loaded.
2009/06/01 15:24:39|         0 Objects expired.
2009/06/01 15:24:39|    160638 Objects cancelled.
2009/06/01 15:24:39|     46804 Duplicate URLs purged.
2009/06/01 15:24:39|         0 Swapfile clashes avoided.
2009/06/01 15:24:39|   Took 2.74 seconds (277631.37 objects/sec).
2009/06/01 15:24:39| Beginning Validation Procedure
2009/06/01 15:24:39|   786432 Entries Validated so far.
2009/06/01 15:24:40|   Completed Validation Procedure
2009/06/01 15:24:40|   Validated 1430597 Entries
2009/06/01 15:24:40|   store_swap_size = 12295252
2009/06/01 15:24:40| storeLateRelease: released 0 objects
2009/06/01 15:24:40| squidaio_queue_request: WARNING - Queue congestion

Thanks,

Quin



----- Original Message ----
From: Quin Guin <[email protected]>
To: squid-dev <[email protected]>
Sent: Wednesday, June 3, 2009 7:51:47 AM
Subject: Re: squid  3.1 is restarting multiple times a day!

I know I should have included this in my origanl post about these resets..


Squid Cache: Version 3.1.0.8-20090531
configure options:  '--prefix=/usr/local/squid-3.1.0.8' '--with-pthreads' 
'--enable-storeio=ufs,aufs' '--enable-fd-config' '--with-filedescriptors=16384' 
'--enable-removal-policies=heap,lru' '--disable-dependency-tracking' 
'--disable-ident-lookups' '--disable-wccp' '--disable-wccpv2' 
'--enable-follow-x-forwarded-for' '--disable-htcp' 
--with-squid=/usr/local/src/dist/SQUID/squid-3.1.0.8-20090531 
--enable-ltdl-convenience



----- Original Message ----
From: Quin Guin <[email protected]>
To: squid-dev <[email protected]>
Sent: Wednesday, June 3, 2009 7:30:33 AM
Subject: squid  3.1 is restarting multiple times a day!

Hi,

I installed 3.1 and it is working well but it is restarting itself multiple 
times a day 4+ times. No core dump is being generated but I do see this in the 
cache log every time it restarts.

2009/06/03 00:22:28| assertion failed: MemBuf.cc:391: "new_cap > (size_t) 
capacity"
2009/06/03 00:22:32| Starting Squid Cache version 3.1.0.8-20090531 for 
x86_64-unknown-linux-gnu...

I would like to do what I can to help resolve this issue because  3.1 is 
working well and performance is good but I would like to resolve this issue 
before installing 3.1 onto more servers. Here are the performance numbers for 
this install:


Version 3.1.0.8-20090531 Quad-Core CPU Intel(R) Xeon(R) CPU L5420 @ 2.50GHz RAM 8 GB HDD 3x SAS,Fujitsu,147Gb,15K OS RHEL4 AS U7 64bit – 2.6.9-78.0.13.ELsmp Users 6 RPS 286.42 Request Hit Ratio 48.42%, 47.8% CPU Usage, 5 minute avg: 16.12% CPU Usage, 60 minute avg: 16.46%

Thanks,

Quin


Your issue appears to be this open bug:
http://www.squid-cache.org/bugs/show_bug.cgi?id=2674

The missing DNS error message is another minor bug. But nothing to seriously worry about.

Amos
--
Please be using
  Current Stable Squid 2.7.STABLE6 or 3.0.STABLE15
  Current Beta Squid 3.1.0.8 or 3.0.STABLE16-RC1

Reply via email to