[squid-users] unexpected debug output

senor frio_cervesa at hotmail.com
Thu Nov 17 07:15:29 UTC 2016


I discovered that 'squid -k rotate' toggles cache.log output into full debug mode as if I had done 'squid -k debug'.  Execute a second rotate and it toggles debug off. This only happens when I have an ecap adapter configured. Comment out those lines and everything works as expected.

My question is about the debug behavior. If this isn't a bug, I'd like to understand the reason rotate would do this. There are no errors when parsing or in normal operation. The ecap adapter works except for a memory leak. Someone is looking into that. I've found it necessary to execute a reconfigure to ensure debugging is off.


squid.conf has 'debug_options ALL,1'.

'squid -k reconfigure' corrects the logging.

Tried with logfile_rotate 0, 1 and 10.

Rotating the log just ends up filling the next log to max and squid halts.

I've run 3.5.19 with same results. Will be trying 22 when I get a minute.

Any ideas are appreciated.


Version and options:

squid -v
Squid Cache: Version 3.5.17
Service Name: squid
configure options:  '--enable-ecap' '--disable-arch-native' '--with-openssl' '--enable-ssl' '--enable-ltdl-convenience' '--enable-linux-netfilter' '--enable-auth' '--with-libcap' '--with-default-user=squid' '--sysconfdir=/etc/squid' '--with-logdir=/var/log/squid' '--with-swapdir=/var/spool/squid' '--enable-wccpv2' '--enable-ssl-crtd'


I don't see anything interesting in these first logged lines after executing 'squid -k rotate':

2016/11/16 21:42:22| Set Current Directory to /var/spool/squid
2016/11/16 21:42:22.952 kid1| storeDirWriteCleanLogs: Starting...
2016/11/16 21:42:22.953 kid1| 6,5| disk.cc(71) file_open: file_open: FD 10
2016/11/16 21:42:22.953 kid1| 51,3| fd.cc(198) fd_open: fd_open() FD 10 /var/spool/squid/swap.state.clean
2016/11/16 21:42:22.953 kid1| 47,3| ufs/UFSSwapDir.cc(978) writeCleanStart: opened /var/spool/squid/swap.state.clean, FD 10
2016/11/16 21:42:22.953 kid1| 6,5| disk.cc(126) file_close: file_close: FD 228 really closing

2016/11/16 21:42:22.953 kid1| 51,3| fd.cc(93) fd_close: fd_close FD 228 /var/spool/squid/swap.state
2016/11/16 21:42:22.953 kid1| 5,5| ModEpoll.cc(116) SetSelect: FD 228, type=1, handler=0, client_data=0, timeout=0
2016/11/16 21:42:22.953 kid1| 5,5| ModEpoll.cc(116) SetSelect: FD 228, type=2, handler=0, client_data=0, timeout=0
2016/11/16 21:42:22.953 kid1| 47,3| ufs/UFSSwapDir.cc(753) closeLog: Cache Dir #0 log closed on FD 228
2016/11/16 21:42:22.953 kid1| 21,2| disk.cc(503) xrename: xrename: renaming /var/spool/squid/swap.state.clean to /var/spool/squid/swap.state
2016/11/16 21:42:22.953 kid1| 6,5| disk.cc(71) file_open: file_open: FD 12
2016/11/16 21:42:22.953 kid1| 51,3| fd.cc(198) fd_open: fd_open() FD 12 /var/spool/squid/swap.state.last-clean
2016/11/16 21:42:22.954 kid1| 6,5| disk.cc(126) file_close: file_close: FD 12 really closing

2016/11/16 21:42:22.954 kid1| 51,3| fd.cc(93) fd_close: fd_close FD 12 /var/spool/squid/swap.state.last-clean
2016/11/16 21:42:22.954 kid1| 5,5| ModEpoll.cc(116) SetSelect: FD 12, type=1, handler=0, client_data=0, timeout=0
2016/11/16 21:42:22.954 kid1| 5,5| ModEpoll.cc(116) SetSelect: FD 12, type=2, handler=0, client_data=0, timeout=0
2016/11/16 21:42:22.954 kid1| 6,5| disk.cc(126) file_close: file_close: FD 10 really closing

2016/11/16 21:42:22.954 kid1| 51,3| fd.cc(93) fd_close: fd_close FD 10 /var/spool/squid/swap.state.clean
2016/11/16 21:42:22.954 kid1| 5,5| ModEpoll.cc(116) SetSelect: FD 10, type=1, handler=0, client_data=0, timeout=0
2016/11/16 21:42:22.954 kid1| 5,5| ModEpoll.cc(116) SetSelect: FD 10, type=2, handler=0, client_data=0, timeout=0
2016/11/16 21:42:22.954 kid1| 6,5| disk.cc(71) file_open: file_open: FD 10
2016/11/16 21:42:22.954 kid1| 51,3| fd.cc(198) fd_open: fd_open() FD 10 /var/spool/squid/swap.state
2016/11/16 21:42:22.954 kid1| 50,3| ufs/UFSSwapDir.cc(735) openLog: Cache Dir #0 log opened on FD 10
2016/11/16 21:42:22.954 kid1|   Finished.  Wrote 3677 entries.

Thanks for any help.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.squid-cache.org/pipermail/squid-users/attachments/20161117/1c2948fd/attachment.html>


More information about the squid-users mailing list