Re: [ilugd] NFS client - Urgent
Yashpal Nagar wrote: The above problem appears intermittently and would disappear if i just umount and mount the file systems manually. Do a dmesg on you NFS-server and see if you are seeing any hdd issues on the server. -- raj shekhar http://rajshekhar.net I've never made anyone's life easier and you know it! Read the latest at my blog: Happy Holi at Stanford, California, USA! http://rajshekhar.net/blog/archives/341-Happy-Holi-at-Stanford,-California,-USA!.html ___ ilugd mailinglist -- ilugd@lists.linux-delhi.org http://frodo.hserus.net/mailman/listinfo/ilugd Archives at: http://news.gmane.org/gmane.user-groups.linux.delhi http://www.mail-archive.com/ilugd@lists.linux-delhi.org/
[ilugd] NFS client - Urgent
Hi All We are running NFS clients on Redhat and SUSE linux distros with the following options. NFSSERVER:/home/idpssupp/REPORTS/mnt/amosoutput nfs nfsvers=3,bg,intr 0 0 for some time quite happily, recently we have encountered the messages on NFS clients Apr 7 07:11:51 NFSCLIENT kernel: nfs: server NFSSERVER OK Apr 7 07:11:51 NFSCLIENT kernel: nfs_statfs: statfs error = 116 Apr 7 07:21:06 NFSCLIENT kernel: nfs_statfs: statfs error = 116 I have tried using... NFSSERVER:/home/idpssupp/REPORTS/mnt/amosoutput nfs nfsvers=3,hard,intr,noac 0 0 and ran mount -o remount /mnt/amosoutput but still we get the following... NFSCLIENT:~ # ls -l /mnt/amosoutput /bin/ls: /mnt/amosoutput: Stale NFS file handle The above problem appears intermittently and would disappear if i just umount and mount the file systems manually. A call is already raised with Redhat but still no response. We have mix of NFS clients on RHEL ES 4, SLES 9 and NFS server is AIX 5.2 ML10. Any help is much appreciated. Regards, Yash ___ ilugd mailinglist -- ilugd@lists.linux-delhi.org http://frodo.hserus.net/mailman/listinfo/ilugd Archives at: http://news.gmane.org/gmane.user-groups.linux.delhi http://www.mail-archive.com/ilugd@lists.linux-delhi.org/
Re: [ilugd] NFS client - Urgent
We are running NFS clients on Redhat and SUSE linux distros with the following options. NFSSERVER:/home/idpssupp/REPORTS/mnt/amosoutput nfs nfsvers=3,bg,intr 0 0 for some time quite happily, recently we have encountered the messages on NFS clients Apr 7 07:11:51 NFSCLIENT kernel: nfs: server NFSSERVER OK Apr 7 07:11:51 NFSCLIENT kernel: nfs_statfs: statfs error = 116 Apr 7 07:21:06 NFSCLIENT kernel: nfs_statfs: statfs error = 116 2009/4/8 Justin Moore jmo...@sagisys.com: You are losing connection to the NFS server. Is the connection stable? Have you tried mounting hard,nointr? To clear the stale file handle warnings you can remove the nfs filesystem from the /etc/mtab and remount. You need to recheck all hardware cables and switches.If it was working okay then you should not loose it. Also check disk transfer rates with hdparm. Check if memory is being reported correctly with free -- Regards, Sudev Barar Read http://blog.sudev.in for topics ranging from here to there. PS: I know most of people do not follow email niceties (mostly they are not aware) but if you follow bottom post/in-line post style of email conversations it becomes a whole lot easier to carry on meaningful dialogue and you can snip out what is not meaningful too. Most people just hit reply button and top post leaving prior message appended uselessly at bottom. See if you can adopt this style and persuade others. In case you are already doing this . great, spread the message. ___ ilugd mailinglist -- ilugd@lists.linux-delhi.org http://frodo.hserus.net/mailman/listinfo/ilugd Archives at: http://news.gmane.org/gmane.user-groups.linux.delhi http://www.mail-archive.com/ilugd@lists.linux-delhi.org/
Re: [ilugd] NFS client - Urgent
On Wed, Apr 8, 2009 at 5:17 PM, Yashpal Nagar yashna...@gmail.com wrote: Hi All We are running NFS clients on Redhat and SUSE linux distros with the following options. NFSSERVER:/home/idpssupp/REPORTS/mnt/amosoutput nfs nfsvers=3,bg,intr 0 0 for some time quite happily, recently we have encountered the messages on NFS clients Apr 7 07:11:51 NFSCLIENT kernel: nfs: server NFSSERVER OK Apr 7 07:11:51 NFSCLIENT kernel: nfs_statfs: statfs error = 116 Apr 7 07:21:06 NFSCLIENT kernel: nfs_statfs: statfs error = 116 I have tried using... NFSSERVER:/home/idpssupp/REPORTS/mnt/amosoutput nfs nfsvers=3,hard,intr,noac 0 0 and ran mount -o remount /mnt/amosoutput but still we get the following... NFSCLIENT:~ # ls -l /mnt/amosoutput /bin/ls: /mnt/amosoutput: Stale NFS file handle The above problem appears intermittently and would disappear if i just umount and mount the file systems manually. A call is already raised with Redhat but still no response. We have mix of NFS clients on RHEL ES 4, SLES 9 and NFS server is AIX 5.2 ML10. Any help is much appreciated. Regards, Yash ___ ilugd mailinglist -- ilugd@lists.linux-delhi.org http://frodo.hserus.net/mailman/listinfo/ilugd Archives at: http://news.gmane.org/gmane.user-groups.linux.delhi http://www.mail-archive.com/ilugd@lists.linux-delhi.org/ 1. Check Network connection between NFS server and clients for TX/RX errors. 2. Are you using automount to mount NFS if yes , check timeout period 3. Check by changing the NFS parameter sync to async, 4. Increase the number of running nfsd by modify /etc/sysconfig/nfs RPCNFSDCOUNT=16 (default is 8) Then restart nfsd. 5. check dmesg for NFS error ___ ilugd mailinglist -- ilugd@lists.linux-delhi.org http://frodo.hserus.net/mailman/listinfo/ilugd Archives at: http://news.gmane.org/gmane.user-groups.linux.delhi http://www.mail-archive.com/ilugd@lists.linux-delhi.org/
Re: [ilugd] NFS client - Urgent
On Wed, Apr 8, 2009 at 12:57 PM, Justin Moore jmo...@sagisys.com wrote: You are losing connection to the NFS server. Is the connection stable? Have you tried mounting hard,nointr? To clear the stale file handle warnings you can remove the nfs filesystem from the /etc/mtab and remount. Justin Removed the entry from /etc/mtab and ran the mount -o remount /mnt/amosoutput I really doubt if the above works, it simply come to next prompt I still see NFSCLIENT:~ # ls -l /mnt/amosoutput /bin/ls: /mnt/amosoutput: Stale NFS file handle the entry for the /mnt/amosoutput didn't reappear in /etc/mtab after I ran the mount with remount option. mount though shows the new options picked with for the shares in question. nointr -- I can try with, but why even after remount, it doesn't pick options. Regards, Yash ___ ilugd mailinglist -- ilugd@lists.linux-delhi.org http://frodo.hserus.net/mailman/listinfo/ilugd Archives at: http://news.gmane.org/gmane.user-groups.linux.delhi http://www.mail-archive.com/ilugd@lists.linux-delhi.org/
Re: [ilugd] NFS client - Urgent
On Wed, Apr 8, 2009 at 1:02 PM, Sudev Barar sba...@gmail.com wrote: We are running NFS clients on Redhat and SUSE linux distros with the following options. NFSSERVER:/home/idpssupp/REPORTS/mnt/amosoutput nfs nfsvers=3,bg,intr 0 0 for some time quite happily, recently we have encountered the messages on NFS clients Apr 7 07:11:51 NFSCLIENT kernel: nfs: server NFSSERVER OK Apr 7 07:11:51 NFSCLIENT kernel: nfs_statfs: statfs error = 116 Apr 7 07:21:06 NFSCLIENT kernel: nfs_statfs: statfs error = 116 2009/4/8 Justin Moore jmo...@sagisys.com: You are losing connection to the NFS server. Is the connection stable? Have you tried mounting hard,nointr? To clear the stale file handle warnings you can remove the nfs filesystem from the /etc/mtab and remount. You need to recheck all hardware cables and switches.If it was working okay then you should not loose it. Also check disk transfer rates with hdparm. Check if memory is being reported correctly with free Both (NFS server, NFS client) servers are 1000mbps duplex full. hdparm on NFS client you mean? NFSclient server is 15k RPM disks (2 T) storage, RAID 1 RAID5 with two raid array controllers. Regards, Yash ___ ilugd mailinglist -- ilugd@lists.linux-delhi.org http://frodo.hserus.net/mailman/listinfo/ilugd Archives at: http://news.gmane.org/gmane.user-groups.linux.delhi http://www.mail-archive.com/ilugd@lists.linux-delhi.org/
Re: [ilugd] NFS client - Urgent
On Wed, April 8, 2009 5:53 pm, Yashpal Nagar wrote: On Wed, Apr 8, 2009 at 1:02 PM, Sudev Barar sba...@gmail.com wrote: You need to recheck all hardware cables and switches.If it was working okay then you should not loose it. Also check disk transfer rates with hdparm. Check if memory is being reported correctly with free Both (NFS server, NFS client) servers are 1000mbps duplex full. hdparm on NFS client you mean? NFSclient server is 15k RPM disks (2 T) storage, RAID 1 RAID5 with two raid array controllers. Have you checked the network connectivity? Use ping/ping -f and see if there are any dropped packets. Maybe the network card is giving trouble. Regards Vivek Kapoor http://exain.com ___ ilugd mailinglist -- ilugd@lists.linux-delhi.org http://frodo.hserus.net/mailman/listinfo/ilugd Archives at: http://news.gmane.org/gmane.user-groups.linux.delhi http://www.mail-archive.com/ilugd@lists.linux-delhi.org/
Re: [ilugd] NFS client - Urgent
You are losing connection to the NFS server. Is the connection stable? Have you tried mounting hard,nointr? To clear the stale file handle warnings you can remove the nfs filesystem from the /etc/mtab and remount. Justin On Wed, 8 Apr 2009 7:47 am, Yashpal Nagar wrote: Hi All We are running NFS clients on Redhat and SUSE linux distros with the following options. NFSSERVER:/home/idpssupp/REPORTS/mnt/amosoutput nfs nfsvers=3,bg,intr 0 0 for some time quite happily, recently we have encountered the messages on NFS clients Apr 7 07:11:51 NFSCLIENT kernel: nfs: server NFSSERVER OK Apr 7 07:11:51 NFSCLIENT kernel: nfs_statfs: statfs error = 116 Apr 7 07:21:06 NFSCLIENT kernel: nfs_statfs: statfs error = 116 I have tried using... NFSSERVER:/home/idpssupp/REPORTS/mnt/amosoutput nfs nfsvers=3,hard,intr,noac 0 0 and ran mount -o remount /mnt/amosoutput but still we get the following... NFSCLIENT:~ # ls -l /mnt/amosoutput /bin/ls: /mnt/amosoutput: Stale NFS file handle The above problem appears intermittently and would disappear if i just umount and mount the file systems manually. A call is already raised with Redhat but still no response. We have mix of NFS clients on RHEL ES 4, SLES 9 and NFS server is AIX 5.2 ML10. Any help is much appreciated. Regards, Yash ___ ilugd mailinglist -- ilugd@lists.linux-delhi.org http://frodo.hserus.net/mailman/listinfo/ilugd Archives at: http://news.gmane.org/gmane.user-groups.linux.delhi http://www.mail-archive.com/ilugd@lists.linux-delhi.org/ -- Justin Moore SagiSys LLC (877)394-4838 ext. 101 (Toll Free) (727)362-1423 (Fax) http://www.SagiSys.com ___ ilugd mailinglist -- ilugd@lists.linux-delhi.org http://frodo.hserus.net/mailman/listinfo/ilugd Archives at: http://news.gmane.org/gmane.user-groups.linux.delhi http://www.mail-archive.com/ilugd@lists.linux-delhi.org/
Re: [ilugd] NFS client - Urgent
.2009/4/8 Yashpal Nagar yashna...@gmail.com: You need to recheck all hardware cables and switches.If it was working okay then you should not loose it. Also check disk transfer rates with hdparm. Check if memory is being reported correctly with free Both (NFS server, NFS client) servers are 1000mbps duplex full. hdparm on NFS client you mean? NFSclient server is 15k RPM disks (2 T) storage, RAID 1 RAID5 with two raid array controllers. hdparm -tT /dev/md0 or what ever. -- Regards, Sudev Barar Read http://blog.sudev.in for topics ranging from here to there. PS: I know most of people do not follow email niceties (mostly they are not aware) but if you follow bottom post/in-line post style of email conversations it becomes a whole lot easier to carry on meaningful dialogue and you can snip out what is not meaningful too. Most people just hit reply button and top post leaving prior message appended uselessly at bottom. See if you can adopt this style and persuade others. In case you are already doing this . great, spread the message. ___ ilugd mailinglist -- ilugd@lists.linux-delhi.org http://frodo.hserus.net/mailman/listinfo/ilugd Archives at: http://news.gmane.org/gmane.user-groups.linux.delhi http://www.mail-archive.com/ilugd@lists.linux-delhi.org/