Le samedi 24 avril 2010 02:09:58, Amos Jeffries a écrit : > Luis Daniel Lucio Quiroz wrote: > > Le samedi 24 avril 2010 00:12:09, Amos Jeffries a écrit : > >> Luis Daniel Lucio Quiroz wrote: > >>> 1272079797.597 52 192.168.203.19 TCP_MEM_HIT/200 4845 GET > >>> http://fernanda.okay.com.mx:3128/squid-internal-periodic/store_digest - > >>> NONE/- application/cache-digest > >>> > >>> is this good. I mean my caches are HITTING store_digest. > >>> > >>> Or shall i add an acl to discart cache hits in this class of request? > >> > >> This is one cache fetching the store index from a peer. > >> > >> Amos > > > > Yes i understand that > > i'm concerning about TCP_MEM_HIT, this hit could block the other caché to > > get the fresh copy of digest and instead that squid is storing a cache > > one? > > > > I mean, in place with activity in where objects goes in and out, when > > other cache request the digest, it could get a non--fresh info. > > Oh. I'm thinking its just because the digests was stored in the RAM > cache at the time. Digest is only recreated every N minutes. Fetches are > served form wherever the latest one was stored. > > Amos ha ok, so this warants me i'm getting the right diggest
