Not sure if this is the proper list, but I've spent days browsing for the solution. This only happens on the CentOS machine, so I'll start here.
After upgrading CPAN, I get the following:
undefined subroutine &Compress::Zlib::gzopen called at /usr/lib/perl5/5.8.5/CPAN/Tarzip.pm line 102.
"gzopen" seems to refer to a PHP command. Am I missing a module, or have I hit a bug?
KS
On 2/13/07, FTN ftnx@ksbase.com wrote:
After upgrading CPAN, I get the following: undefined subroutine &Compress::Zlib::gzopen called at /usr/lib/perl5/5.8.5/CPAN/Tarzip.pm line 102.
"gzopen" seems to refer to a PHP command. Am I missing a module, or have I hit a bug?
CPAN on rpm based machines is EVIL in so many ways it's not even funny.
To start, nothing installed via cpan identifies itself to the rpm database, so anything you install which depends on a perl (CPAN) module will fail because of missing dependencies. If you install something via CPAN and upgrade the perl rpm for some reason (a perl upgrade is released which translates regex into plain understandable english perhaps) it's entirely possible that it will overwrite things installed via cpan. This will cause you headaches down the road. In addition, if a cpan module gets overly greedy with its dependency build, you can end up with a broken or otherwise non-working perl install.
It's much better to use the perl modules in the rpmforge and c.k.o repositories, or cpan2rpm, or cpanflute. These will not interfere in such a manner.
</soap box>
Guys:
On a 4.4 box I use as my desktop at work, I just did a "yum update" today for the first time in a while and it got the latest kernel 2.6.9-42.0.8. Upon rebooting afterward nfs now fails to mount two nfs shares on another Linux (very old Red Hat 6.2) box, that it always had mounted previously.
google didn't help me much, and I couldn't find anything abouu it in the Centos forums (fora??).
I'm getting an error something like "RPC error: Program not registered."
did the new kernel break NFS on us?
Thanks!
fredex wrote:
Guys:
On a 4.4 box I use as my desktop at work, I just did a "yum update" today for the first time in a while and it got the latest kernel 2.6.9-42.0.8. Upon rebooting afterward nfs now fails to mount two nfs shares on another Linux (very old Red Hat 6.2) box, that it always had mounted previously.
google didn't help me much, and I couldn't find anything abouu it in the Centos forums (fora??).
I'm getting an error something like "RPC error: Program not registered."
did the new kernel break NFS on us?
Thanks!
Fred, Don't know if this will help you but I had something similar happen. Adding the following:
nfsd /proc/fs/nfsd nfsd auto,defaults 0 0
to /etc/fstab and issuing a 'mount -a' fixed my problem.
http://www.centos.org/modules/newbb/viewtopic.php?topic_id=6891&forum=30
Kirk Bocek
On Tue, Feb 13, 2007 at 04:40:43PM -0800, Kirk Bocek wrote:
fredex wrote:
Guys:
On a 4.4 box I use as my desktop at work, I just did a "yum update" today for the first time in a while and it got the latest kernel 2.6.9-42.0.8. Upon rebooting afterward nfs now fails to mount two nfs shares on another Linux (very old Red Hat 6.2) box, that it always had mounted previously.
google didn't help me much, and I couldn't find anything abouu it in the Centos forums (fora??).
I'm getting an error something like "RPC error: Program not registered."
did the new kernel break NFS on us?
Thanks!
Fred, Don't know if this will help you but I had something similar happen. Adding the following:
nfsd /proc/fs/nfsd nfsd auto,defaults 0 0
to /etc/fstab and issuing a 'mount -a' fixed my problem.
http://www.centos.org/modules/newbb/viewtopic.php?topic_id=6891&forum=30
Kirk Bocek
Kirk:
Thanks for the info. I'll give it a try when I get back to the office.
On Tue, Feb 13, 2007 at 04:40:43PM -0800, Kirk Bocek wrote:
fredex wrote:
Guys:
On a 4.4 box I use as my desktop at work, I just did a "yum update" today for the first time in a while and it got the latest kernel 2.6.9-42.0.8. Upon rebooting afterward nfs now fails to mount two nfs shares on another Linux (very old Red Hat 6.2) box, that it always had mounted previously.
google didn't help me much, and I couldn't find anything abouu it in the Centos forums (fora??).
I'm getting an error something like "RPC error: Program not registered."
did the new kernel break NFS on us?
Thanks!
Fred, Don't know if this will help you but I had something similar happen. Adding the following:
nfsd /proc/fs/nfsd nfsd auto,defaults 0 0
to /etc/fstab and issuing a 'mount -a' fixed my problem.
http://www.centos.org/modules/newbb/viewtopic.php?topic_id=6891&forum=30
Kirk Bocek
Nope. No change. Nice try, though.
Fred
On Wed, 2007-02-14 at 08:11 -0500, fredex wrote:
On Tue, Feb 13, 2007 at 04:40:43PM -0800, Kirk Bocek wrote:
fredex wrote:
Guys:
On a 4.4 box I use as my desktop at work, I just did a "yum update" today for the first time in a while and it got the latest kernel 2.6.9-42.0.8. Upon rebooting afterward nfs now fails to mount two nfs shares on another Linux (very old Red Hat 6.2) box, that it always had mounted previously.
google didn't help me much, and I couldn't find anything abouu it in the Centos forums (fora??).
I'm getting an error something like "RPC error: Program not registered."
did the new kernel break NFS on us?
Thanks!
Fred, Don't know if this will help you but I had something similar happen. Adding the following:
nfsd /proc/fs/nfsd nfsd auto,defaults 0 0
to /etc/fstab and issuing a 'mount -a' fixed my problem.
http://www.centos.org/modules/newbb/viewtopic.php?topic_id=6891&forum=30
Kirk Bocek
Nope. No change. Nice try, though.
hmmmm
how about this kind of entry in /etc/fstab:
10.2.0.2:/home/sun_vms /mnt/xeon nfs defaults 0 0
1st column is the machine name(or IP), a colon, and directory for the nfs server export ... 2nd column is the place you want to mount it on this machine, 3rd columns is the file type (nfs) ... from that point what you already have is OK (for columns 4,5,6).
Thanks, Johnny Hughes
On Wed, 2007-02-14 at 08:35 -0600, Johnny Hughes wrote:
On Wed, 2007-02-14 at 08:11 -0500, fredex wrote:
On Tue, Feb 13, 2007 at 04:40:43PM -0800, Kirk Bocek wrote:
fredex wrote:
Guys:
On a 4.4 box I use as my desktop at work, I just did a "yum update" today for the first time in a while and it got the latest kernel 2.6.9-42.0.8. Upon rebooting afterward nfs now fails to mount two nfs shares on another Linux (very old Red Hat 6.2) box, that it always had mounted previously.
google didn't help me much, and I couldn't find anything abouu it in the Centos forums (fora??).
I'm getting an error something like "RPC error: Program not registered."
did the new kernel break NFS on us?
Thanks!
Fred, Don't know if this will help you but I had something similar happen. Adding the following:
nfsd /proc/fs/nfsd nfsd auto,defaults 0 0
to /etc/fstab and issuing a 'mount -a' fixed my problem.
http://www.centos.org/modules/newbb/viewtopic.php?topic_id=6891&forum=30
Kirk Bocek
Nope. No change. Nice try, though.
hmmmm
how about this kind of entry in /etc/fstab:
10.2.0.2:/home/sun_vms /mnt/xeon nfs defaults 0 0
1st column is the machine name(or IP), a colon, and directory for the nfs server export ... 2nd column is the place you want to mount it on this machine, 3rd columns is the file type (nfs) ... from that point what you already have is OK (for columns 4,5,6).
Thanks, Johnny Hughes _______________________________________________ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
http://nfs.sourceforge.net/nfs-howto/ar01s07.html
HTH
Calin
================================================= Only adults have difficulty with childproof caps.
On Wed, Feb 14, 2007 at 05:23:40PM +0200, kalinix wrote:
On Wed, 2007-02-14 at 08:35 -0600, Johnny Hughes wrote:
On Wed, 2007-02-14 at 08:11 -0500, fredex wrote:
On Tue, Feb 13, 2007 at 04:40:43PM -0800, Kirk Bocek wrote:
fredex wrote:
Guys:
On a 4.4 box I use as my desktop at work, I just did a "yum update" today for the first time in a while and it got the latest kernel 2.6.9-42.0.8. Upon rebooting afterward nfs now fails to mount two nfs shares on another Linux (very old Red Hat 6.2) box, that it always had mounted previously.
google didn't help me much, and I couldn't find anything abouu it in the Centos forums (fora??).
I'm getting an error something like "RPC error: Program not registered."
did the new kernel break NFS on us?
Thanks!
Fred, Don't know if this will help you but I had something similar happen. Adding the following:
nfsd /proc/fs/nfsd nfsd auto,defaults 0 0
to /etc/fstab and issuing a 'mount -a' fixed my problem.
http://www.centos.org/modules/newbb/viewtopic.php?topic_id=6891&forum=30
Kirk Bocek
Nope. No change. Nice try, though.
hmmmm
how about this kind of entry in /etc/fstab:
10.2.0.2:/home/sun_vms /mnt/xeon nfs defaults 0 0
1st column is the machine name(or IP), a colon, and directory for the nfs server export ... 2nd column is the place you want to mount it on this machine, 3rd columns is the file type (nfs) ... from that point what you already have is OK (for columns 4,5,6).
Thanks, Johnny Hughes _______________________________________________ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
http://nfs.sourceforge.net/nfs-howto/ar01s07.html
HTH
Calin
Calin:
Thanks for the reply. I've already checked that, and it looks to me as if everything that should be running, is running.
here's rpcinfo -p on the NFS server machine: # rpcinfo -p program vers proto port 100000 2 tcp 111 portmapper 100000 2 udp 111 portmapper 100005 1 udp 674 mountd 100005 1 tcp 676 mountd 100005 2 udp 679 mountd 100005 2 tcp 681 mountd
and here is the same thing on the client where nfs mounts aren't working (BTW, they ARE working from other clients, still):
# rpcinfo -p 128.2.2.25 program vers proto port 100000 2 tcp 111 portmapper 100000 2 udp 111 portmapper 100005 1 udp 674 mountd 100005 1 tcp 676 mountd 100005 2 udp 679 mountd 100005 2 tcp 681 mountd
one of my co-workers who uses fc4 on his desktop was able to mount it on his client, and it is still mounted from other (non-linux) systems in the server room, so it looks to me like a problem on the client machine. since the problem appeared only after the latest kernel update, it seems likely to be related to that.
Thanks for your assistance!
Fred
On Wed, Feb 14, 2007 at 10:49:16AM -0500, fredex enlightened us:
one of my co-workers who uses fc4 on his desktop was able to mount it on his client, and it is still mounted from other (non-linux) systems in the server room, so it looks to me like a problem on the client machine. since the problem appeared only after the latest kernel update, it seems likely to be related to that.
Have you rebooted into an older kernel to confirm this is in fact the case?
Matt
On Thu, Feb 15, 2007 at 11:41:32AM -0500, Matt Hyclak wrote:
On Wed, Feb 14, 2007 at 10:49:16AM -0500, fredex enlightened us:
one of my co-workers who uses fc4 on his desktop was able to mount it on his client, and it is still mounted from other (non-linux) systems in the server room, so it looks to me like a problem on the client machine. since the problem appeared only after the latest kernel update, it seems likely to be related to that.
Have you rebooted into an older kernel to confirm this is in fact the case?
Not yet. that machine, besides being my linux desktop, serves up several vmware images too and it's a pain to shut 'em all back down. but I will be doing it before too long just to make sure.
On Thu, Feb 15, 2007 at 11:41:32AM -0500, Matt Hyclak wrote:
On Wed, Feb 14, 2007 at 10:49:16AM -0500, fredex enlightened us:
one of my co-workers who uses fc4 on his desktop was able to mount it on his client, and it is still mounted from other (non-linux) systems in the server room, so it looks to me like a problem on the client machine. since the problem appeared only after the latest kernel update, it seems likely to be related to that.
Have you rebooted into an older kernel to confirm this is in fact the case?
Matt
Well, now it's my turn to be embarrassed :(
Rebooting the older kernel didn't change things at all, NFS still wasn't working.
So off to the machine serving the nfs shares to reboot it (it had been up for a good six months anyway, so a reboot wasn't necessarily a bad idea, just to let it fsck the partitions). Once it came up I could then mount the nfs shares without issue.
sorry about all the whining gang, but I appreciate the attempts to help!
On Wed, Feb 14, 2007 at 10:49:16AM -0500, fredex wrote:
program vers proto port 100000 2 tcp 111 portmapper 100000 2 udp 111 portmapper 100005 1 udp 674 mountd 100005 1 tcp 676 mountd 100005 2 udp 679 mountd 100005 2 tcp 681 mountd
No nfsd? Huh. Also I see only v1 and v2, no v3 services. This is what I see on my NFS server:
% rpcinfo -p SERVER | egrep '(nfs|mountd)' 100003 2 udp 2049 nfs 100003 3 udp 2049 nfs 100003 2 tcp 2049 nfs 100003 3 tcp 2049 nfs 100005 1 udp 661 mountd 100005 1 tcp 664 mountd 100005 2 udp 661 mountd 100005 2 tcp 664 mountd 100005 3 udp 661 mountd 100005 3 tcp 664 mountd
Anyway, no v3 services may mean you need to force a v2 mount so try adding nfsvers=2 to the options field in fstab. It's possible the new kernel is defaulting to v3 mounts. Dunno. Just a guess :-)
On Wed, Feb 14, 2007 at 08:35:43AM -0600, Johnny Hughes wrote:
On Wed, 2007-02-14 at 08:11 -0500, fredex wrote:
On Tue, Feb 13, 2007 at 04:40:43PM -0800, Kirk Bocek wrote:
fredex wrote:
Guys:
On a 4.4 box I use as my desktop at work, I just did a "yum update" today for the first time in a while and it got the latest kernel 2.6.9-42.0.8. Upon rebooting afterward nfs now fails to mount two nfs shares on another Linux (very old Red Hat 6.2) box, that it always had mounted previously.
google didn't help me much, and I couldn't find anything abouu it in the Centos forums (fora??).
I'm getting an error something like "RPC error: Program not registered."
did the new kernel break NFS on us?
Thanks!
Fred, Don't know if this will help you but I had something similar happen. Adding the following:
nfsd /proc/fs/nfsd nfsd auto,defaults 0 0
to /etc/fstab and issuing a 'mount -a' fixed my problem.
http://www.centos.org/modules/newbb/viewtopic.php?topic_id=6891&forum=30
Kirk Bocek
Nope. No change. Nice try, though.
hmmmm
how about this kind of entry in /etc/fstab:
10.2.0.2:/home/sun_vms /mnt/xeon nfs defaults 0 0
1st column is the machine name(or IP), a colon, and directory for the nfs server export ... 2nd column is the place you want to mount it on this machine, 3rd columns is the file type (nfs) ... from that point what you already have is OK (for columns 4,5,6).
Thanks, Johnny Hughes
Johnny:
Thanks for the reply!
I've already got this entry in /etc/fstab:
128.2.2.25:/mnt/cisrc /usr3 nfs rw,rsize=8192,wsize=8192,hard,intr,user 0 0
and it looks to me as if it should be adequate,... at least it always HAS been up until the kernel update I got yesterday.
The error occurs during boot when the startup tries to mount all filesystems, as well as whenever I type: "mount /usr3" or anything else that causes the sys to try mounting it.
Any other ideas?
Thanks again!
Fred
PS: yes, nfs is running:
# pwd /etc/rc.d/init.d # ./nfs status Shutting down NFS mountd: rpc.mountd (pid 4816) is running... nfsd (pid 4812 4811 4810 4809 4806 4805 4804 4803) is running... rpc.rquotad (pid 4799) is running... #
also, I note that (as shown above) the nfs status command says its shutting down NFS mountd. I believe this is an error in the nfs script, which contains:
status) [ -x /usr/sbin/rpc.svcgssd ] && /sbin/service rpcsvcgssd status echo -n $"Shutting down NFS mountd: " status rpc.mountd status nfsd if [ -n "$RQUOTAD" -a "$RQUOTAD" != "no" ]; then status rpc.rquotad fi ;;
Do you agree? Should a bug be filed on this?
fredex wrote:
On Tue, Feb 13, 2007 at 04:40:43PM -0800, Kirk Bocek wrote:
fredex wrote:
Guys:
On a 4.4 box I use as my desktop at work, I just did a "yum update" today for the first time in a while and it got the latest kernel 2.6.9-42.0.8. Upon rebooting afterward nfs now fails to mount two nfs shares on another Linux (very old Red Hat 6.2) box, that it always had mounted previously.
google didn't help me much, and I couldn't find anything abouu it in the Centos forums (fora??).
I'm getting an error something like "RPC error: Program not registered."
did the new kernel break NFS on us?
Thanks!
Fred, Don't know if this will help you but I had something similar happen. Adding the following:
nfsd /proc/fs/nfsd nfsd auto,defaults 0 0
to /etc/fstab and issuing a 'mount -a' fixed my problem.
http://www.centos.org/modules/newbb/viewtopic.php?topic_id=6891&forum=30
Kirk Bocek
Nope. No change. Nice try, though.
Forgot to mention that the /etc/fstab entry goes on the server and not the client.
Kirk Bocek
On Wed, Feb 14, 2007 at 09:47:45AM -0800, Kirk Bocek wrote:
fredex wrote:
On Tue, Feb 13, 2007 at 04:40:43PM -0800, Kirk Bocek wrote:
fredex wrote:
Guys:
On a 4.4 box I use as my desktop at work, I just did a "yum update" today for the first time in a while and it got the latest kernel 2.6.9-42.0.8. Upon rebooting afterward nfs now fails to mount two nfs shares on another Linux (very old Red Hat 6.2) box, that it always had mounted previously.
google didn't help me much, and I couldn't find anything abouu it in the Centos forums (fora??).
I'm getting an error something like "RPC error: Program not registered."
did the new kernel break NFS on us?
Thanks!
Fred, Don't know if this will help you but I had something similar happen. Adding the following:
nfsd /proc/fs/nfsd nfsd auto,defaults 0 0
to /etc/fstab and issuing a 'mount -a' fixed my problem.
http://www.centos.org/modules/newbb/viewtopic.php?topic_id=6891&forum=30
Kirk Bocek
Nope. No change. Nice try, though.
Forgot to mention that the /etc/fstab entry goes on the server and not the client.
Kirk Bocek
I can try it, but... it's been working for years, and STILL works with other clients, so it would seem pretty odd if this helps.
On Wed, Feb 14, 2007 at 01:06:49PM -0500, fredex wrote:
On Wed, Feb 14, 2007 at 09:47:45AM -0800, Kirk Bocek wrote:
fredex wrote:
On Tue, Feb 13, 2007 at 04:40:43PM -0800, Kirk Bocek wrote:
fredex wrote:
Guys:
On a 4.4 box I use as my desktop at work, I just did a "yum update" today for the first time in a while and it got the latest kernel 2.6.9-42.0.8. Upon rebooting afterward nfs now fails to mount two nfs shares on another Linux (very old Red Hat 6.2) box, that it always had mounted previously.
google didn't help me much, and I couldn't find anything abouu it in the Centos forums (fora??).
I'm getting an error something like "RPC error: Program not registered."
did the new kernel break NFS on us?
Thanks!
Fred, Don't know if this will help you but I had something similar happen. Adding the following:
nfsd /proc/fs/nfsd nfsd auto,defaults 0 0
to /etc/fstab and issuing a 'mount -a' fixed my problem.
http://www.centos.org/modules/newbb/viewtopic.php?topic_id=6891&forum=30
Kirk Bocek
Nope. No change. Nice try, though.
Forgot to mention that the /etc/fstab entry goes on the server and not the client.
Kirk Bocek
I can try it, but... it's been working for years, and STILL works with other clients, so it would seem pretty odd if this helps.
Nope, no difference.
fredex wrote:
I can try it, but... it's been working for years, and STILL works with other clients, so it would seem pretty odd if this helps.
Nope, no difference.
Well I offered my fix because my setup had been working for years too.
Kirk Bocek
On Wed, Feb 14, 2007 at 10:17:04AM -0800, Kirk Bocek wrote:
fredex wrote:
I can try it, but... it's been working for years, and STILL works with other clients, so it would seem pretty odd if this helps.
Nope, no difference.
Well I offered my fix because my setup had been working for years too.
Kirk Bocek
thanks for trying to help, I do appreciate it.
-
fredex -
FTN -
Jim Perrin -
Johnny Hughes -
kalinix -
Kirk Bocek -
Matt Hyclak -
Stephen Harris