Hi,
Just upgraded a remote 4.1 server [basically Samba] with yum update. Went through the process and every showed OK, when I did a restart, the only thing working is ping. I could ping it, but that's it.
Could not connect anymore through SSH, could not connect through SMB. My smb.conf is still there and it still shows my settings. Testparm -L reports the correct info, but nobody can connect.
I now have the server being shipped back. Anybody also having problems?
On Fri, 2006-03-24 at 12:41 -0800, centos@911networks.com wrote:
Hi,
Just upgraded a remote 4.1 server [basically Samba] with yum update. Went through the process and every showed OK, when I did a restart, the only thing working is ping. I could ping it, but that's it.
Could not connect anymore through SSH, could not connect through SMB. My smb.conf is still there and it still shows my settings. Testparm -L reports the correct info, but nobody can connect.
I now have the server being shipped back. Anybody also having problems?
Mine is not a server, but it is operational in a private network. Just today (about and hour ago) did the upgrade, However, I was at latest 4.2 update and don't run Samba. Worked like a charm.
I'm using it now for this and other activities.
My only extra repo item is an rpmforge for liblame. All else is CentOS base... um I do have an mplayer installed to, almost forgot that.
I hope you have good luck with it. I also hope that you find some valuable clues or lessons that help in the future.
If there is anything I can do (check my logs, ID diffs in how we did it, ...) that might help your investigation, let me know.
Bill
On Fri, 2006-03-24 at 12:41 -0800, centos@911networks.com wrote:
Hi,
Just upgraded a remote 4.1 server [basically Samba] with yum update. Went through the process and every showed OK, when I did a restart, the only thing working is ping. I could ping it, but that's it.
Could not connect anymore through SSH, could not connect through SMB. My smb.conf is still there and it still shows my settings. Testparm -L reports the correct info, but nobody can connect.
I now have the server being shipped back. Anybody also having problems?
I have upgraded dozens of machines with no problems ... I have seen one other person with ssh type issues.
I would say when you get the server, take a look at any .rpmnew items on the date of install and see what they might say. You might some special configurations that might need some newer items rolled into them.
Also, if you have selinux on, take a look at that very hard.
Johnny Hughes wrote:
I would say when you get the server, take a look at any .rpmnew items on the date of install and see what they might say. You might some special configurations that might need some newer items rolled into them.
Will do
Also, if you have selinux on, take a look at that very hard.
No SELinux
centos@911networks.com wrote:
Hi,
Just upgraded a remote 4.1 server [basically Samba] with yum update. Went through the process and every showed OK, when I did a restart, the only thing working is ping. I could ping it, but that's it.
Could not connect anymore through SSH, could not connect through SMB. My smb.conf is still there and it still shows my settings. Testparm -L reports the correct info, but nobody can connect.
I now have the server being shipped back. Anybody also having problems?
is it possible to have someone look at the console ? its possible that something is waiting for a y/n answer, also - do you have mdadm / dmraid setup on this machine ? I've seen dmraid sometimes cause issues with the newer lvm2 stuff.
Karanbir Singh wrote:
is it possible to have someone look at the console ? its possible that
Yes I did
something is waiting for a y/n answer, also - do you have mdadm / dmraid setup on this machine ? I've seen dmraid sometimes cause issues with the
No. I will get the server back on Monday and will figure it out [hopefully].
I'd really like to know what this turned out to be...
I had an issue a while back where a kernel update caused a server to fail reboot - something about the combination of SCSI, Software RAID, and Grub. I had to run a grub script from the console to get it all back up. (Making an emergency plane ride worthwhile as well as fun!)
The most recent kernel update did not cause any problems.
My advice? Don't host more than about 2-4 hour drive from home...
-Ben
On Friday 24 March 2006 12:41, centos@911networks.com wrote:
Hi,
Just upgraded a remote 4.1 server [basically Samba] with yum update. Went through the process and every showed OK, when I did a restart, the only thing working is ping. I could ping it, but that's it.
Could not connect anymore through SSH, could not connect through SMB. My smb.conf is still there and it still shows my settings. Testparm -L reports the correct info, but nobody can connect.
I now have the server being shipped back. Anybody also having problems?
-- Thanks http://www.911networks.com When the network has to work _______________________________________________ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
-- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean.
Benjamin Smith wrote:
I'd really like to know what this turned out to be...
It was sendmail that would hang during the boot. At that point the startup would wait, wait and wait... I do not need sendmail, so I disabled it from the startup.
My advice? Don't host more than about 2-4 hour drive from home...
Not an option...
centos@911networks.com wrote:
Benjamin Smith wrote:
I'd really like to know what this turned out to be...
It was sendmail that would hang during the boot. At that point the startup would wait, wait and wait... I do not need sendmail, so I disabled it from the startup.
<insert qmail|postfix|courier jab here>
My advice? Don't host more than about 2-4 hour drive from home...
Not an option...
Nor for me, that's why I endeavor to keep a duplicate of my production server locally to try out such things. Of course that doesn't save me from things like fat-fingered iptables scripts. I did so today and my host (Server Beach) was super quick to console in and run the commands I specified.
JT
On Tue, 2006-03-28 at 13:20, centos@911networks.com wrote:
Benjamin Smith wrote:
I'd really like to know what this turned out to be...
It was sendmail that would hang during the boot. At that point the startup would wait, wait and wait... I do not need sendmail, so I disabled it from the startup.
That means it's really DNS that's broken - and/or you don't have an entry for localhost in /etc/hosts. Sendmail likes to be able to resolve names for all of your interfaces so it knows what mail it should accept and it will try pretty hard before giving up. My guess is that you had the caching-nameserver package installed but had local changes to named.conf that where needed to make DNS work. An update to caching-nameserver wipes out local changes.
On Tue, 2006-03-28 at 15:07 -0600, Les Mikesell wrote:
On Tue, 2006-03-28 at 13:20, centos@911networks.com wrote:
Benjamin Smith wrote:
I'd really like to know what this turned out to be...
It was sendmail that would hang during the boot. At that point the startup would wait, wait and wait... I do not need sendmail, so I disabled it from the startup.
That means it's really DNS that's broken - and/or you don't have an entry for localhost in /etc/hosts. Sendmail likes to be able to resolve names for all of your interfaces so it knows what mail it should accept and it will try pretty hard before giving up. My guess is that you had the caching-nameserver package installed but had local changes to named.conf that where needed to make DNS work. An update to caching-nameserver wipes out local changes.
---- I've seen a lot of people completely ignore the stop sign - the top 2 lines of /etc/hosts and hack it away completely and that causes all sorts of issues.
My money is on careless edit of /etc/hosts ;-)
Craig
or have a RAC card of some kind..:)
Benjamin Smith wrote:
I'd really like to know what this turned out to be...
I had an issue a while back where a kernel update caused a server to fail reboot - something about the combination of SCSI, Software RAID, and Grub. I had to run a grub script from the console to get it all back up. (Making an emergency plane ride worthwhile as well as fun!)
The most recent kernel update did not cause any problems.
My advice? Don't host more than about 2-4 hour drive from home...
-Ben
On Friday 24 March 2006 12:41, centos@911networks.com wrote:
Hi,
Just upgraded a remote 4.1 server [basically Samba] with yum update. Went through the process and every showed OK, when I did a restart, the only thing working is ping. I could ping it, but that's it.
Could not connect anymore through SSH, could not connect through SMB. My smb.conf is still there and it still shows my settings. Testparm -L reports the correct info, but nobody can connect.
I now have the server being shipped back. Anybody also having problems?
-- Thanks http://www.911networks.com When the network has to work _______________________________________________ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
-- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean.
-
Benjamin Smith -
centos＠911networks.com -
Craig White -
Johnny Hughes -
JT Justman -
Karanbir Singh -
Les Mikesell -
William L. Maltby -
William Warren