Amos, The problem mentioned before (first e-mail) happens with the squidGuard helper.
Sharing a single cache_dir (diskd) entry for a multiple worker setup is a valid approach, right? Could be related to this setup the error below? kid1| assertion failed: store.cc:1086: "EBIT_TEST(flags, ENTRY_ABORTED)" What could cause this other error? kid1| assertion failed: Read.cc:69: "fd_table[conn->fd].halfClosedReader != NULL" I currently have both fatal errors in my servers. The last one with more occurrences. The servers have a lot of traffic and I'm using two workers. Thanks in advance. William Lima ----- Original Message ----- From: "Amos Jeffries" <[email protected]> To: "William Lima" <[email protected]>, [email protected] Sent: Wednesday, May 24, 2017 10:13:53 AM Subject: Re: [squid-dev] Memory error On 24/05/17 06:00, William Lima wrote: > Hi, > > I'm saw errors like these below: > > 2017/05/23 14:35:34 kid2| helperHandleRead: unexpected reply on channel 0 > from redirector #Hlpr2 '220543.35796293814' > 2017/05/23 14:35:34 kid2| helperHandleRead: unexpected reply on channel 0 > from redirector #Hlpr2 '3380' > 2017/05/23 14:35:34 kid2| helperHandleRead: unexpected reply on channel 0 > from redirector #Hlpr2 '2' > 2017/05/23 14:35:34 kid2| helperHandleRead: unexpected reply on channel 0 > from redirector #Hlpr2 'd' > 2017/05/23 14:35:34 kid2| helperHandleRead: unexpected reply on channel 0 > from redirector #Hlpr2 'null"' > > It seems like a memory corruption, since sometimes the srv->rbuf data shows > some HTML content not related to a valid reply. > I'm currently using squid 3.5.25. > > Any clue? > Any other data available? for example what the helper is being passed by Squid and what it thinks is being sent back to Squid? Amos _______________________________________________ squid-dev mailing list [email protected] http://lists.squid-cache.org/listinfo/squid-dev
