Matus UHLAR - fantomas wrote:
On 28.06 22:54, Hendrik Voigtländer wrote:
This is what I would do:
- deaktivate all cache_dir which are close to 100%
- deaktivate all cache_dirs on one disk
- repartition it with a single partition & mkreiserfs
- put one cache_dir on it
- repeat with the other disks
- if you are on a big uplink hit ratios shouldnt suffer to much, take your time and let squid refill the cache_dirs.
- be careful with the cache_dir size. with your disks not more than 50% and you don't need to keep more than a week of traffic.
well, I would use more than 50%, maybe 70%, or better, ~80% of free space on created filesystem, where the overhead is already substracted
Hi,
In general I agree on a maximum possible usage of ~80%, but if I read the mails correctly she has something about 2GB of RAM but 5 x 73 GB Harddisk for the cache_dirs only. This is about 300GB of storage, what about the RAM requirements when using ~80% ?
http://www.mail-archive.com/[EMAIL PROTECTED]/msg18052.html
IMHO there is not enough RAM for all disks in the machine. As having more disks gives more speed I think reducing the maximum usage is a better option than reducing the number of disks.
Good idea for such a large storage. Any idea how this will influence the RAM requirements?
I would check and probably increase maximum_object_size up to ~32MB
and use cache_replacement_policy heap LFUDA
Regards, Hendrik
