You might try instrumenting nfs/nfs_bio.c to find where the EINTR is coming from.
If it's reproducible then you should be able (with a few reboots) to track it back quite a distance. julian On Mon, 29 Mar 1999 mest...@visi.com wrote: > > Greetings, > > I am running two machines with current from Sunday, March 28 at 3:20PM > CST, but the problems I am seeing I have seen for a while now (about a > month). I thought an upgrade might fix them. Here are the specs on the > machines: > > Machine A: > Dual PPro 200MHz, 128MB RAM, SCSI disks on an onboard Adaptec 7880, rl0 > driver (Realtek 8139) > > Machine B: > Dual Pentium 133, 64MB RAM, IDE disks, xl0 driver (3COM 905B) > > Both machines are connected via a 10/100 hub, and are operating at > 100Mb/s half duplex. Machine A is the NFS server, and machine B is the > client. Machine B runs apache and serves up some web pages over NFS > from machine A. I am mounting with options rw,intr in machine B's > fstab. According to the mount_nfs manpage, I am using nfsv3 since my > server allows that. When certain files on machine A are accessed by the > web daemon on machine B, the httpd process hangs, and when I eventually > use apachectl to stop it, here is what is printed out on machine B's > console: > > Mar 29 18:30:04 wall /kernel: nfs_getpages: error 4 > Mar 29 18:30:04 wall /kernel: vm_fault: pager read error, pid 274 > (httpd) > > I can reproduce this fairly easily if need be. I looked in the mail > archives, but didn't see anyone else seeing this problem recently. Anyone > have any clues on this one? I can provide more information if needed. > Thanks. > > -- > Kyle Mestery > StorageTek's Storage Networking Group > Protect your right to privacy: www.freecrypto.org > > > > To Unsubscribe: send mail to majord...@freebsd.org > with "unsubscribe freebsd-current" in the body of the message > To Unsubscribe: send mail to majord...@freebsd.org with "unsubscribe freebsd-current" in the body of the message
