On 20/10/2016 1:59 a.m., erdosain9 wrote: > Hi. > I try this > http://www.linuxtopic.com/2016/03/clear-squid-cache-and-re-create-cache.html > > but still the same... > Squid auto-corrects the cache data contents. The message you are seeing is what gets logged during the correction process. You do not have to do anything to "fix" the cache data, Squid has already done so when it logged that message. Clearing the cache does not fix anything - simply quietens the logs, since you have erased all the broken objects along with the good ones. The messages *will* occur again whenever the actual problem is encountered next. Until you find out what that is and fix it the log entries will just keep coming back. <snip> > > Is there some problem???? Not in the config file. These errors occur in the following situations: 1) multiple Squid running accessing the same cache_dir path and overwriting each others data. - can also be SMP workers sharing a UFS, AUFS or diskd cache_dir. 2) A Squid crash happening while some files are waiting to be written, but after the swap.state journal has been updated to mention their existence. 3) HDD or SSD disk corruption. Squid pushes disks very hard, and they do die eventually. Usually much faster than manufacturer specs indicate. Thats not a manufacturer issue, just a symptom of how hard Squid pushes the disks. 4) some non-Squid script or software playing around with the Squid cache directory contents. (see #1) > how config diskd??? Replace "aufs" with "diskd" in your cache_dir line(s). Then reconfigure Squid. Amos _______________________________________________ squid-users mailing list squid-users@xxxxxxxxxxxxxxxxxxxxx http://lists.squid-cache.org/listinfo/squid-users
