Hi all,
Am I the only one to have had a totally trouble-free upgrade to 5.2?? I get the feeling that most everybody seems to have *some* problems...
Ha, until now it looks ok, but i only upgraded yesterday evening...
On Wed, 25 Jun 2008 12:29:41 +0200 "Sorin@Gmail" sorin.srbu@gmail.com wrote:
sorin.srbu> Hi all, sorin.srbu> sorin.srbu> Am I the only one to have had a totally trouble-free upgrade to 5.2?? I get the sorin.srbu> feeling that most everybody seems to have *some* problems... sorin.srbu> -- sorin.srbu> Best Wishes sorin.srbu> Sorin sorin.srbu> --------------------------------------------- sorin.srbu> http://home-skynet.servehttp.com/ sorin.srbu> Proud member of TEAM OS/2. sorin.srbu> Mountainbiker [Kona Kilauea - Member of Equipe Les Cafards VTT] sorin.srbu> Motorcyclist [Honda VFR750F-'97] sorin.srbu> MCSE, MCP+I, MCP, A+ [Knowledge is power!] sorin.srbu> --------------------------------------------- sorin.srbu> () ascii ribbon campaign - against html e-mail sorin.srbu> /\ sorin.srbu> sorin.srbu> MotD: sorin.srbu> CentOS: What Sysadmins run at home. sorin.srbu> sorin.srbu> sorin.srbu> sorin.srbu>
Am Mittwoch, 25. Juni 2008 schrieb Sorin@Gmail:
Am I the only one to have had a totally trouble-free upgrade to 5.2?? I get the feeling that most everybody seems to have *some* problems...
Zero problems here after upgrading a bunch of servers. I guess the majority of users without problems is just not as loud. :-)
Thank you CentOS team for the good work!
regards, Andreas Micklei
Sorin@Gmail wrote:
Hi all,
Am I the only one to have had a totally trouble-free upgrade to 5.2?? I get the feeling that most everybody seems to have *some* problems...
well, how it works is that the millions of machines out there where it just worked, wont get reported, but if things break people could use a helpinghand, or just something to sync with other people who have similar problems.
Going by your take of things, looking at the bugs.centos.org site - one might conclude that OhMyGod,NothingWorks,PanicPanicPanic, is there anything on this ship that is good!!
Which would, even if I may say so myself, be quite wrong!
I think everything is good :D Just a few people run into issues, so they report them ( and they should, its what makes us a community! )
Karanbir Singh wrote:
Sorin@Gmail wrote:
Hi all,
Am I the only one to have had a totally trouble-free upgrade to 5.2?? I get the feeling that most everybody seems to have *some* problems...
well, how it works is that the millions of machines out there where it just worked, wont get reported, but if things break people could use a helpinghand, or just something to sync with other people who have similar problems.
Going by your take of things, looking at the bugs.centos.org site - one might conclude that OhMyGod,NothingWorks,PanicPanicPanic, is there anything on this ship that is good!!
Which would, even if I may say so myself, be quite wrong!
I think everything is good :D Just a few people run into issues, so they report them ( and they should, its what makes us a community! )
As KB says, we estimate more than 2 million CentOS installs out there, so the number of major issues are fairly small compared to that.
We have identified several upstream issues that we have put in the Release Notes (see the known issues section) and for most we have testing packages to fix them.
Thanks, Johnny Hughes
Karanbir Singh <> scribbled on Wednesday, June 25, 2008 12:51 PM:
Am I the only one to have had a totally trouble-free upgrade to 5.2?? I get the feeling that most everybody seems to have *some* problems...
well, how it works is that the millions of machines out there where it just worked, wont get reported, but if things break people could use a helpinghand, or just something to sync with other people who have similar problems.
Going by your take of things, looking at the bugs.centos.org site - one might conclude that OhMyGod,NothingWorks,PanicPanicPanic, is there anything on this ship that is good!!
Which would, even if I may say so myself, be quite wrong!
I think everything is good :D Just a few people run into issues, so they report them ( and they should, its what makes us a community! )
You're are right I guess. 8-) Like the other guy said, the ones with problems are just louder than the rest of us. ;-)
Actually I did panic a bit when yum finished looking for updates and reported it needed some 300-400MB. I was afarid I'd hit the max on /. Turns out it worked flawlessly. My mother OTOH, has a 256kbps line down and 300-400M will take a while to download... She doesn't have any space-constraints though on / as I had. Win some, lose some...
All in all, CentOS 5.2 seems to work fine. My web server and sshd is still up and chugging along. I'm happy. 8-)
Sorin@Gmail wrote:
Hi all,
Am I the only one to have had a totally trouble-free upgrade to 5.2?? I get the feeling that most everybody seems to have *some* problems...
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Only problem I had was my dovecot imap server going down, but that was my fault for not reading through the new .conf file properly..
other than that its been very smooth
big thanks to everyone involved, great work!
cheers
Dunc
Dunc wrote:
Sorin@Gmail wrote:
Hi all,
Am I the only one to have had a totally trouble-free upgrade to 5.2?? I get the feeling that most everybody seems to have *some* problems...
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Only problem I had was my dovecot imap server going down, but that was my fault for not reading through the new .conf file properly..
other than that its been very smooth
big thanks to everyone involved, great work!
If there is something in the conf file that needs to change (that worked in 5.1 and does not in 5.2) let us know and we can add it to the release notes and/or wiki.
Johnny Hughes wrote:
If there is something in the conf file that needs to change (that worked in 5.1 and does not in 5.2) let us know and we can add it to the release notes and/or wiki.
It was in the listening section as follows:
# A space separated list of IP or host addresses where to listen in for # connections. "*" listens in all IPv4 interfaces. "[::]" listens in all IPv6 # interfaces, but may also listen in all IPv4 interfaces depending on the # operating system. Use "*, [::]" for listening both IPv4 and IPv6. # # If you want to specify ports for each service, you will need to configure # these settings inside the protocol imap/pop3 { ... } section, so you can # specify different ports for IMAP/POP3. For example: # protocol imap { # listen = *:10143 # ssl_listen = *:10943 # .. # } # protocol pop3 { # listen = *:10100 # .. # } listen = "*" #listen = [::]
The listen bit defaults to [::]. I had to change it to "*" as above otherwise I couldn't connect to it. (connection refused)
As it says above, [::] only listens on IPv4 'depending on the operating system'. So something has changed from 5.1 to 5.2 which means that is no longer the case.
thanks
Dunc
Dunc wrote:
Johnny Hughes wrote:
If there is something in the conf file that needs to change (that worked in 5.1 and does not in 5.2) let us know and we can add it to the release notes and/or wiki.
It was in the listening section as follows:
# A space separated list of IP or host addresses where to listen in for # connections. "*" listens in all IPv4 interfaces. "[::]" listens in all IPv6 # interfaces, but may also listen in all IPv4 interfaces depending on the # operating system. Use "*, [::]" for listening both IPv4 and IPv6. # # If you want to specify ports for each service, you will need to configure # these settings inside the protocol imap/pop3 { ... } section, so you can # specify different ports for IMAP/POP3. For example: # protocol imap { # listen = *:10143 # ssl_listen = *:10943 # .. # } # protocol pop3 { # listen = *:10100 # .. # } listen = "*" #listen = [::]
The listen bit defaults to [::]. I had to change it to "*" as above otherwise I couldn't connect to it. (connection refused)
As it says above, [::] only listens on IPv4 'depending on the operating system'. So something has changed from 5.1 to 5.2 which means that is no longer the case.
thanks
Dunc
That's strange Dunc.
The new config file was written to /etc/dovecot/conf.rpmnew (as expected) for me, so your modified original doesn't get overwritten. I can see no difference between the 5.1 and 5.2 config files wrt the 'listen' directive. Mine was, and still is, set to the default [::] and still works fine. Further, I have IPv6 disabled on the server. I have tested on both i386 and x86_64 platforms and each are working fine both with and without ssl/tls.
Did you restart the dovecot service after it was updated? (I'm not sure if yum update does this automatically or not when updating a daemon service that is running).
Regards,
Ned
Ned Slider wrote:
That's strange Dunc.
The new config file was written to /etc/dovecot/conf.rpmnew (as expected) for me, so your modified original doesn't get overwritten. I can see no difference between the 5.1 and 5.2 config files wrt the 'listen' directive. Mine was, and still is, set to the default [::] and still works fine. Further, I have IPv6 disabled on the server. I have tested on both i386 and x86_64 platforms and each are working fine both with and without ssl/tls.
Did you restart the dovecot service after it was updated? (I'm not sure if yum update does this automatically or not when updating a daemon service that is running).
Regards,
Ned
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Hi
Yes it was written as rpmnew, and indeed the default was the same. But using the original I was using before, changing nothing, with [::] I could not connect after the upgrade. I restarted many times, and eventually changing the original to [*] allowed me to connect.
I then changed to the new conf (after changing a few things like using mysql to authenticate, minimum uid to match my setup), and again could not connect. Changing the new conf to [*] then solved it.
How are you disabling IPv6, maybe that is making the difference?
thanks
Dunc
Dunc wrote:
Hi
Yes it was written as rpmnew, and indeed the default was the same. But using the original I was using before, changing nothing, with [::] I could not connect after the upgrade. I restarted many times, and eventually changing the original to [*] allowed me to connect.
I then changed to the new conf (after changing a few things like using mysql to authenticate, minimum uid to match my setup), and again could not connect. Changing the new conf to [*] then solved it.
How are you disabling IPv6, maybe that is making the difference?
To disable IPv6 as per the Wiki:
http://wiki.centos.org/FAQ/CentOS5?highlight=(ipv6)#head-47912ebdae3b5ac10ff...
Also, see this thread by Lenard:
http://www.centos.org/modules/newbb/viewtopic.php?viewmode=flat&topic_id...
Ned Slider wrote:
Dunc wrote:
Hi
Yes it was written as rpmnew, and indeed the default was the same. But using the original I was using before, changing nothing, with [::] I could not connect after the upgrade. I restarted many times, and eventually changing the original to [*] allowed me to connect.
I then changed to the new conf (after changing a few things like using mysql to authenticate, minimum uid to match my setup), and again could not connect. Changing the new conf to [*] then solved it.
How are you disabling IPv6, maybe that is making the difference?
To disable IPv6 as per the Wiki:
http://wiki.centos.org/FAQ/CentOS5?highlight=(ipv6)#head-47912ebdae3b5ac10ff...
Hi
Seems that is making the difference. With IPv6 disabled both * and :: work. If I enable it again and reboot only * works..
Dunc
Dunc wrote:
Ned Slider wrote:
Dunc wrote:
Hi
Yes it was written as rpmnew, and indeed the default was the same. But using the original I was using before, changing nothing, with [::] I could not connect after the upgrade. I restarted many times, and eventually changing the original to [*] allowed me to connect.
I then changed to the new conf (after changing a few things like using mysql to authenticate, minimum uid to match my setup), and again could not connect. Changing the new conf to [*] then solved it.
How are you disabling IPv6, maybe that is making the difference?
To disable IPv6 as per the Wiki:
http://wiki.centos.org/FAQ/CentOS5?highlight=(ipv6)#head-47912ebdae3b5ac10ff...
Hi
Seems that is making the difference. With IPv6 disabled both * and :: work. If I enable it again and reboot only * works..
Dunc
OK, thanks Dunc.
As this seems to affect the default install configuration we presumably need to add a note somewhere to reflect this. Thanks for spotting and reporting it.
Has anyone else experienced this or can confirm it?
Ned Slider wrote:
As this seems to affect the default install configuration we presumably need to add a note somewhere to reflect this. Thanks for spotting and reporting it.
Has anyone else experienced this or can confirm it?
There are some of us, who really use IPv6 in real life :D
Karanbir Singh <> scribbled on Thursday, June 26, 2008 1:31 PM:
Ned Slider wrote:
As this seems to affect the default install configuration we presumably need to add a note somewhere to reflect this. Thanks for spotting and reporting it.
Has anyone else experienced this or can confirm it?
There are some of us, who really use IPv6 in real life :D
Somebody's actually using IPv6? One comes to think of flying pigs for some reason... ;-)
Hi,
On Wed, 2008-06-25 at 12:29 +0200, Sorin@Gmail wrote:
Hi all,
Am I the only one to have had a totally trouble-free upgrade to 5.2?? I get the feeling that most everybody seems to have *some* problems...
I upgraded 4 machines (all virtual) without any problems from 5.1 to 5.2, tonight I'll upgrade the host machine but I do not expect any problems. I'm only cautious to upgrade my old webserver which is a dual PIII and is located about 100km from my home ;) I wouldn't like a kernel panic there.
Regards,
Michel
On Wed, 2008-06-25 at 14:39 +0200, Michel van Deventer wrote:
Hi,
<snip>
I upgraded 4 machines (all virtual) without any problems from 5.1 to 5.2, tonight I'll upgrade the host machine but I do not expect any problems. I'm only cautious to upgrade my old webserver which is a dual PIII and is located about 100km from my home ;) I wouldn't like a kernel panic there.
Suggestion: before rebooting, edit grub's config to default to the old kernel. Then reboot, selecting the new one. If it fails, when the machine is booted again, it will be in the old stuff. If it works, you can then edit grub's config again to default to the new one.
Regards,
Michel
<snip>
HTH
on 6-25-2008 3:29 AM Sorin@Gmail spake the following:
Hi all,
Am I the only one to have had a totally trouble-free upgrade to 5.2?? I get the feeling that most everybody seems to have *some* problems...
Only a few problems so far on the list. A small percentage of the total user base. But many people will not upgrade on the first day, or even the first week.
On Wed, Jun 25, 2008 at 2:14 PM, Scott Silva ssilva@sgvwater.com wrote:
on 6-25-2008 3:29 AM Sorin@Gmail spake the following:
Hi all,
Am I the only one to have had a totally trouble-free upgrade to 5.2?? I get the feeling that most everybody seems to have *some* problems...
Only a few problems so far on the list. A small percentage of the total
user base. But many people will not upgrade on the first day, or even the first week.
On my Desktop box, I had a problem, after powering down the box, where it would not boot into the new kernel. Adding "acpi=off" to the line for the new kernel eliminated that problem. On my daughters Desktop box, there were no issues. We are the pioneers and others who follow us will learn from the issues that we had. :-)
-
Andreas Micklei -
Dunc -
Johnny Hughes -
Karanbir Singh -
Lanny Marcus -
Michel van Deventer -
Ned Slider -
Scott Silva -
Sorin＠Gmail -
Test -
William L. Maltby