Any idea what might be causing this? We saw some fileservers sucking large amounts of memory, and then this result in afsmonitor, note the huge number of connections (numclients). There was also a warning in the logs: Sep 4 14:15:47 afs2 fileserver[1008]: ****Allocated more volume headers, probably leak**** Any idea what might be causing this? Only thing special about these servers is that I have them configured with relatively large inode-max (512000) and file-max (256000) and ulimit -HSn set to 16384. Restarting the fileservers caused the performance issue and memory utilization to drop back to something reasonable. -- Nathan "Uetrecht, Daniel J." wrote: > > This looks to correspond to what I saw shutting down. > > -----Original Message----- > From: System Administrator [mailto:[EMAIL PROTECTED]] > Sent: Tuesday, September 04, 2001 2:52 PM > To: [EMAIL PROTECTED] > Subject: > > AFSMonitor [Version 1.0] [File Servers, > p. 1 of 1, c. 64 of 271] > > <<< 6 File Servers monitored, 0 alerts on 0 machines > >>> > > host host host > HostsIn NumClients Client FetchData > FetchData FetchData > DiffNetwo* Blocks epoch ops > ops_ok sum (sec) > > afs2 9 1600 22 999631227 31639 > 31581 407.083495 > > afs4 20 793734 10874 991398052 39018849 > 38963461 461001.237 > > 474 > afs7 14 3051 51 977516865 19589224 > 19588905 446837.444 > > 119 > afs8 20 1053856 14437 991399813 54032721 > 54027254 1111331.85 > > 790740342* > afs9 13 578829 7930 992269066 23813006 > 23493202 305443.127 > > 247 > afs10 1 22481 308 999352689 2202910 > 2202797 56034.3538 > > 52 -- ------------------------------------------------------------ Nathan Neulinger EMail: [EMAIL PROTECTED] University of Missouri - Rolla Phone: (573) 341-4841 CIS - Systems Programming Fax: (573) 341-4216 _______________________________________________ OpenAFS-devel mailing list [EMAIL PROTECTED] https://lists.openafs.org/mailman/listinfo/openafs-devel
