These are typical on Solaris. Don't know about other platforms though. I believe they're caused in part by users aborting requests (i.e. hitting the stop button). I get lot's of them.
-Mike On Mon, 31 Mar 2003, Jacobi Michael CRPH wrote: > I have been seeing similar things in the last few days (starting on Friday) on my > 2.5Stable1-20021231 system (Redhat Linux 6.2, Kernel 2.2.25). One comment though - > I have recently upgraded the kernel to 2.2.25 from 2.2.14-7. I was NOT seeing this > before I upgraded the kernel! > > Mike Jacobi > > > -----Original Message----- > From: Hector Miranda [mailto:[EMAIL PROTECTED] > Sent: Monday, March 31, 2003 5:30 PM > To: '[EMAIL PROTECTED]' > Subject: [squid-users] cache.log showing this... > > > Hi fellows... > V880, Ultra SPARC III, Solaris 8, Squid 2.5STABLE1 > Giga Ethernet access to internet, four thousand users behind my proxy and > the cache.log shows this over and over again (1 or 2 times per minute): > > comm_accept: FD 7: (130) Software caused connection abort > httpAccept: FD 7: accept failure: (130) Software caused connection abort > > we thougth there was a problem with the communication speed to the internet, > so the Client's Web browser wold abort the connection, but the down and up > load rates to the internet on my net are about 100-200 Kb/s (through squid), > so this is not the problem... why is this happening?? > > Any ideas? > > Thanks in advance! > > .H�ctor M. > ** *** ****** Mike Diggins **** **** ******** Analyst, Networks ** ******* ** ** Computing and Information Services ** ** ** ** ** McMaster University, Hamilton, Ontario ** ** ** ** (905) 525-9140 Ext. 27471 ABB131A ** ** ********* Fax: (905) 528-3773 *** *** ******** Email: [EMAIL PROTECTED]
