Hi, 2010/8/18 Amos Jeffries <squid3@xxxxxxxxxxxxx>: >>> cache_dir aufs /cache 756842 60 100 > > > Whats missing appears to be min-size=1048576 on the AUFS to push all the > small objects into the better COSS directories. (NOTE: the value is COSS > max-size+1) Duh, now I changed it to: cache_dir aufs /cache 756842 60 100 min-size=1048576 cache_dir coss /coss1 65520 max-size=1048575 max-stripe-waste=32768 block-size=4096 membufs=15 cache_dir coss /coss2 65520 max-size=1048575 max-stripe-waste=32768 block-size=4096 membufs=15 cache_dir coss /coss3 65520 max-size=1048575 max-stripe-waste=32768 block-size=4096 membufs=15 >>> Is there a way to know which cache_dir is the problem and what I can >>> so this doesn't happen? > > AIO is the I/O method preferred by AUFS. That aufs dir is also listed > first, which may affect the default choice. Really? This is only before I had the min-size option, right? Or even with that option you think I should put the aufs cache_dir below the coss cache_dir ? > *49* of 10 million objects are in-transit? that is very low. Though it > could be a result of the queue overload. :-) I got these after the support team called me to check the machine. They had already taken squid down flushing the ebtables rules...so that number shouldn't be right after all. Thank you. - Robert
