| Summary: | squid does not start if /var/spool/squid is empty | ||
|---|---|---|---|
| Product: | [Fedora] Fedora | Reporter: | Michael Weidner <micha> |
| Component: | squid | Assignee: | Jiri Skala <jskala> |
| Status: | CLOSED ERRATA | QA Contact: | Fedora Extras Quality Assurance <extras-qa> |
| Severity: | unspecified | Docs Contact: | |
| Priority: | unspecified | ||
| Version: | 16 | CC: | aglotov, henrik, jonathansteffan, jskala, marcosfrm, mluscon |
| Target Milestone: | --- | ||
| Target Release: | --- | ||
| Hardware: | i686 | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Fixed In Version: | squid-3.2.0.13-4.fc16 | Doc Type: | Bug Fix |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2011-11-14 22:24:14 UTC | Type: | --- |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
|
Description
Michael Weidner
2011-10-18 20:06:51 UTC
There is missing ExecStartPre option in the squid.service file that will call script invoking squid -z as was done in sysv initscript. I'v tested it locally on F16 virtual machine. I'll commit it in next days (still watching BZ#7471{10,25}) squid-3.2.0.13-2.fc16 has been submitted as an update for Fedora 16. https://admin.fedoraproject.org/updates/squid-3.2.0.13-2.fc16 I can confirm the bug is fixed with squid-3.2.0.13-3.fc16 Package squid-3.2.0.13-3.fc16: * should fix your issue, * was pushed to the Fedora 16 testing repository, * should be available at your local mirror within two days. Update it with: # su -c 'yum update --enablerepo=updates-testing squid-3.2.0.13-3.fc16' as soon as you are able to. Please go to the following url: https://admin.fedoraproject.org/updates/FEDORA-2011-14951 then log in and leave karma (feedback). Package squid-3.2.0.13-4.fc16: * should fix your issue, * was pushed to the Fedora 16 testing repository, * should be available at your local mirror within two days. Update it with: # su -c 'yum update --enablerepo=updates-testing squid-3.2.0.13-4.fc16' as soon as you are able to. Please go to the following url: https://admin.fedoraproject.org/updates/FEDORA-2011-14951 then log in and leave karma (feedback). squid-3.2.0.13-4.fc16 has been pushed to the Fedora 16 stable repository. If problems still persist, please make note of it in this bug report. I'm having this bug on F17, squid-3.2.0.16-1.fc17.x86_64.
Sep 03 10:47:28 XXX (squid-1)[12203]: Failed to verify one of the swap directories, Check cache.log
Sep 03 10:47:28 XXX squid[12188]: Squid Parent: (squid-1) process 12203 exited with status 1
Sep 03 10:47:28 XXX squid[12188]: Squid Parent: (squid-1) process 12203 will not be restarted due to repeated, frequent failures
Sep 03 10:47:28 XXX squid[12188]: Exiting due to repeated, frequent failures
Sep 03 10:47:28 XXX squid[12206]: squid: ERROR: Could not send signal 15 to process 1436: (3) No such process
cache.log
2012/09/03 10:47:28 kid1| Starting Squid Cache version 3.2.0.16 for x86_64-redhat-linux-gnu...
2012/09/03 10:47:28 kid1| Process ID 12203
2012/09/03 10:47:28 kid1| Process Roles: worker
2012/09/03 10:47:28 kid1| With 16384 file descriptors available
2012/09/03 10:47:28 kid1| Initializing IP Cache...
2012/09/03 10:47:28 kid1| DNS Socket created at [::], FD 7
2012/09/03 10:47:28 kid1| DNS Socket created at 0.0.0.0, FD 8
2012/09/03 10:47:28 kid1| Adding nameserver XXX.XXX.XXX.XXX from /etc/resolv.conf
2012/09/03 10:47:28 kid1| Adding nameserver XXX.XXX.XXX.XXX from /etc/resolv.conf
2012/09/03 10:47:28 kid1| Logfile: opening log daemon:/var/log/squid/access.log
2012/09/03 10:47:28 kid1| Logfile Daemon: opening log /var/log/squid/access.log
2012/09/03 10:47:28 kid1| Local cache digest enabled; rebuild/rewrite every 3600/3600 sec
2012/09/03 10:47:28 kid1| Store logging disabled
2012/09/03 10:47:28 kid1| Swap maxSize 20971520 + 262144 KB, estimated 1633358 objects
2012/09/03 10:47:28 kid1| Target number of buckets: 81667
2012/09/03 10:47:28 kid1| Using 131072 Store buckets
2012/09/03 10:47:28 kid1| Max Mem size: 262144 KB
2012/09/03 10:47:28 kid1| Max Swap size: 20971520 KB
2012/09/03 10:47:28 kid1| /var/spool/squid/00: (2) No such file or directory
FATAL: Failed to verify one of the swap directories, Check cache.log
for details. Run 'squid -z' to create swap directories
if needed, or if running Squid for the first time.
Squid Cache (Version 3.2.0.16): Terminated abnormally.
CPU Usage: 0.056 seconds = 0.030 user + 0.026 sys
Maximum Resident Size: 53408 KB
Page faults with physical i/o: 0
/var/log/squid/squid.out is empty.
/usr/libexec/squid/cache_swap.sh is failing (or not being executed) somehow.
OK, cache_swap.sh was failing because systemd's ExecStartPre doesn't allow forking processes. Squid needs the "-N" (no daemon mode) there. Err, the "-N" (no daemon mode) *option* there. Please fix cache_swap.sh adding "-N" option to squid. The current F17 package (and probably F18/rawhide) is kind of broken as it will refuse to start in a clean install. Also take a look at Bug 854356 Thanks. (In reply to comment #10) > Please fix cache_swap.sh adding "-N" option to squid. The current F17 > package (and probably F18/rawhide) is kind of broken as it will refuse to > start in a clean install. > > Also take a look at Bug 854356 > > Thanks. First of all I believe there should be opened new bug. I went through the initial squid's code and I guess forking should be invoked after handling swap dirs. I don't see necessity forking due to this operation because the squid is terminated just after swap dirs creation. Henrik, what's your opinion? Done: Bug 859393 |