Hi guys,
Sometime, when we apply a change on squid module (Access Control or Ports) happens a strange problem: squid process is restarted and store rebuilding happens.
The problem when the squid cache dir size increases (in my case i have 5 cache dir in /var/spool/squid3/ directory): a very high time do rebuild caches cause a Terminated abnormally of squid process.
O believe that problem is: squid restart process instead of giving a "-k reconfigure", forcing a store rebuilding:
cache.log:
==================
2015/04/17 09:39:48| Done reading /var/spool/squid3/cache5 swaplog (1864 entries)
2015/04/17 09:39:48| Store rebuilding is 14.83% complete
2015/04/17 09:40:37| Done reading /var/spool/squid3/cache1 swaplog (27647 entries)
2015/04/17 09:40:38| Done reading /var/spool/squid3/cache3 swaplog (27772 entries)
2015/04/17 09:40:38| Done reading /var/spool/squid3/cache2 swaplog (27982 entries)
2015/04/17 09:40:38| Done reading /var/spool/squid3/cache4 swaplog (27967 entries)
2015/04/17 09:40:38| Finished rebuilding storage from disk.
2015/04/17 09:40:38| 112998 Entries scanned
2015/04/17 09:40:38| 0 Invalid entries.
2015/04/17 09:40:38| 0 With invalid flags.
2015/04/17 09:40:38| 112782 Objects loaded.
2015/04/17 09:40:38| 0 Objects expired.
2015/04/17 09:40:38| 216 Objects cancelled.
2015/04/17 09:40:38| 18 Duplicate URLs purged.
2015/04/17 09:40:38| 0 Swapfile clashes avoided.
2015/04/17 09:40:38| Took 50.62 seconds (2227.89 objects/sec).
2015/04/17 09:40:38| Beginning Validation Procedure
2015/04/17 09:40:38| Completed Validation Procedure
2015/04/17 09:40:38| Validated 221055 Entries
2015/04/17 09:40:38| store_swap_size = 17731496256
FATAL: xcalloc: Unable to allocate 4073701254 blocks of 1 bytes!
Squid Cache (Version 3.1.20): Terminated abnormally.
CPU Usage: 50.167 seconds = 49.843 user + 0.324 sys
Maximum Resident Size: 232144 KB
Page faults with physical i/o: 0
Memory usage for squid via mallinfo():
total space in arena: 26868 KB
Ordinary blocks: 26227 KB 7 blks
Small blocks: 0 KB 0 blks
Holding blocks: 53556 KB 15 blks
Free Small blocks: 0 KB
Free Ordinary blocks: 640 KB
Total in use: 79783 KB 297%
Total free: 640 KB 2%
2015/04/17 09:40:42| Starting Squid Cache version 3.1.20 for x86_64-pc-linux-gnu...
2015/04/17 09:40:42| Process ID 42815
2015/04/17 09:40:42| With 65535 file descriptors available
2015/04/17 09:40:42| Initializing IP Cache...
===============
Have a workaround to this?
Thanks!
In the Squid module of Webmin, click on the Module Config link and see what the "Command to apply changes" is set to.
Dear Jamie,
I'm change from:
service squid3 restart
To:
service squid3 reload
And now the cache log don't show directory rebuild!
Thanks
On 21-04-2015 01:25, Jamie Cameron wrote:
Related
Bugs:
#4588Thanks - I will have Webmin use that command in future by default.
On 21/Apr/2015 00:08 Elias Andrade monesa@users.sf.net wrote ..
Dear Jamie,
I'm change from: service squid3 restart
To: service squid3 reload
And now the cache log don't show directory rebuild!
Thanks
On 21-04-2015 01:25, Jamie Cameron wrote:
In the Squid module of Webmin, click on the Module Config link and see what the "Command to apply changes" is set to.
[bugs:#4588] http://sourceforge.net/p/webadmin/bugs/4588 Squid problem when cache is large
Status: open Group: 1.710 Labels: squid Created: Mon Apr 20, 2015 04:51 PM UTC by Elias Andrade Last Updated: Mon Apr 20, 2015 04:51 PM UTC Owner: nobody
Hi guys,
Sometime, when we apply a change on squid module (Access Control or Ports) happens a strange problem: squid process is restarted and store rebuilding happens.
The problem when the squid cache dir size increases (in my case i have 5 cache dir in /var/spool/squid3/ directory): a very high time do rebuild caches cause a Terminated abnormally of squid process.
O believe that problem is: squid restart process instead of giving a "-k reconfigure", forcing a store rebuilding:
cache.log:
================== 2015/04/17 09:39:48| Done reading /var/spool/squid3/cache5 swaplog (1864 entries) 2015/04/17 09:39:48| Store rebuilding is 14.83% complete 2015/04/17 09:40:37| Done reading /var/spool/squid3/cache1 swaplog (27647 entries) 2015/04/17 09:40:38| Done reading /var/spool/squid3/cache3 swaplog (27772 entries) 2015/04/17 09:40:38| Done reading /var/spool/squid3/cache2 swaplog (27982 entries) 2015/04/17 09:40:38| Done reading /var/spool/squid3/cache4 swaplog (27967 entries) 2015/04/17 09:40:38| Finished rebuilding storage from disk. 2015/04/17 09:40:38| 112998 Entries scanned 2015/04/17 09:40:38| 0 Invalid entries. 2015/04/17 09:40:38| 0 With invalid flags. 2015/04/17 09:40:38| 112782 Objects loaded. 2015/04/17 09:40:38| 0 Objects expired. 2015/04/17 09:40:38| 216 Objects cancelled. 2015/04/17 09:40:38| 18 Duplicate URLs purged. 2015/04/17 09:40:38| 0 Swapfile clashes avoided. 2015/04/17 09:40:38| Took 50.62 seconds (2227.89 objects/sec). 2015/04/17 09:40:38| Begin
ning Validation Procedure 2015/04/17 09:40:38| Completed Validation Procedure 2015/04/17 09:40:38| Validated 221055 Entries 2015/04/17 09:40:38| store_swap_size = 17731496256 FATAL: xcalloc: Unable to allocate 4073701254 blocks of 1 bytes!
Squid Cache (Version 3.1.20): Terminated abnormally. CPU Usage: 50.167 seconds = 49.843 user + 0.324 sys Maximum Resident Size: 232144 KB Page faults with physical i/o: 0 Memory usage for squid via mallinfo(): total space in arena: 26868 KB Ordinary blocks: 26227 KB 7 blks Small blocks: 0 KB 0 blks Holding blocks: 53556 KB 15 blks Free Small blocks: 0 KB Free Ordinary blocks: 640 KB Total in use: 79783 KB 297% Total free: 640 KB 2% 2015/04/17 09:40:42| Starting Squid Cache version 3.1.20 for x86_64-pc-linux-gnu... 2015/04/17 09:40:42| Process ID 42815 2015/04/17 09:40:42| With 65535 file descriptors available 2015/04/17 09:40:42| Initializing IP Cache...
===============
Have a workaround to this?
Thanks!
Sent from sourceforge.net because you indicated interest in https://sourceforge.net/p/webadmin/bugs/4588/ https://sourceforge.net/p/webadmin/bugs/4588
To unsubscribe from further messages, please visit https://sourceforge.net/auth/subscriptions/ https://sourceforge.net/auth/subscriptions
[bugs:#4588] Squid problem when cache is large
Status: open Group: 1.710 Labels: squid Created: Mon Apr 20, 2015 04:51 PM UTC by Elias Andrade Last Updated: Tue Apr 21, 2015 06:56 AM UTC Owner: nobody
Hi guys,
Sometime, when we apply a change on squid module (Access Control or Ports) happens a strange problem: squid process is restarted and store rebuilding happens.
The problem when the squid cache dir size increases (in my case i have 5 cache dir in /var/spool/squid3/ directory): a very high time do rebuild caches cause a Terminated abnormally of squid process.
O believe that problem is: squid restart process instead of giving a "-k reconfigure", forcing a store rebuilding:
cache.log:
================== 2015/04/17 09:39:48| Done reading /var/spool/squid3/cache5 swaplog (1864 entries) 2015/04/17 09:39:48| Store rebuilding is 14.83% complete 2015/04/17 09:40:37| Done reading /var/spool/squid3/cache1 swaplog (27647 entries) 2015/04/17 09:40:38| Done reading /var/spool/squid3/cache3 swaplog (27772 entries) 2015/04/17 09:40:38| Done reading /var/spool/squid3/cache2 swaplog (27982 entries) 2015/04/17 09:40:38| Done reading /var/spool/squid3/cache4 swaplog (27967 entries) 2015/04/17 09:40:38| Finished rebuilding storage from disk. 2015/04/17 09:40:38| 112998 Entries scanned 2015/04/17 09:40:38| 0 Invalid entries. 2015/04/17 09:40:38| 0 With invalid flags. 2015/04/17 09:40:38| 112782 Objects loaded. 2015/04/17 09:40:38| 0 Objects expired. 2015/04/17 09:40:38| 216 Objects cancelled. 2015/04/17 09:40:38| 18 Duplicate URLs purged. 2015/04/17 09:40:38| 0 Swapfile clashes avoided. 2015/04/17 09:40:38| Took 50.62 seconds (2227.89 objects/sec). 2015/04/17 09:40:38| Begin
ning Validation Procedure 2015/04/17 09:40:38| Completed Validation Procedure 2015/04/17 09:40:38| Validated 221055 Entries 2015/04/17 09:40:38| store_swap_size = 17731496256 FATAL: xcalloc: Unable to allocate 4073701254 blocks of 1 bytes!
Squid Cache (Version 3.1.20): Terminated abnormally. CPU Usage: 50.167 seconds = 49.843 user + 0.324 sys Maximum Resident Size: 232144 KB Page faults with physical i/o: 0 Memory usage for squid via mallinfo(): total space in arena: 26868 KB Ordinary blocks: 26227 KB 7 blks Small blocks: 0 KB 0 blks Holding blocks: 53556 KB 15 blks Free Small blocks: 0 KB Free Ordinary blocks: 640 KB Total in use: 79783 KB 297% Total free: 640 KB 2% 2015/04/17 09:40:42| Starting Squid Cache version 3.1.20 for x86_64-pc-linux-gnu... 2015/04/17 09:40:42| Process ID 42815 2015/04/17 09:40:42| With 65535 file descriptors available 2015/04/17 09:40:42| Initializing IP Cache...
===============
Have a workaround to this?
Thanks!
Sent from sourceforge.net because you indicated interest in https://sourceforge.net/p/webadmin/bugs/4588/
To unsubscribe from further messages, please visit https://sourceforge.net/auth/subscriptions/
Related
Bugs:
#4588Dear Jamie,
I'm change from:
service squid3 restart
To:
service squid3 reload
And now the cache log don't show directory rebuild!
Thanks
I will have Webmin use "service squid3 reload" in future by default to apply settings, which should avoid this .