I installed CentOS 5.3 on Oct 27, 2009 and immediately updated it to 5.4.
I had not paid attention to this boot log until yesterday when I decided to take a look to do some trouble shooting.
Oct 27 13:16:24 localhost NET[3797]: /sbin/dhclient-script : updated
/etc/resolv.conf
Oct 27 13:16:29 localhost NET[4059]:
/etc/sysconfig/network-scripts/ifup-post : updated /etc/resolv.conf
This entry was probably made right after the fresh installation.
I've done searches on the Internet for answers for this problem and the only thing I've found is that I am not the only one with this problem. But no body has offered any definite solution for it. I even posted the same question in the CentOS forum and haven't got any reply.
From what I have read, this problem may be caused by a bug or bugs in CentOS
5.x, but I hope it is not because I am in the process of switching from Fedora 10 to CentOS 5.4 and I don't want to get disappointed. Trivial as this may be, this does not give me confidence in CentOS distribution.
Can anyone come up with a solution?
Rod Rook wrote:
I installed CentOS 5.3 on Oct 27, 2009 and immediately updated it to 5.4.
I had not paid attention to this boot log until yesterday when I decided to take a look to do some trouble shooting.
Oct 27 13:16:24 localhost NET[3797]: /sbin/dhclient-script : updated
/etc/resolv.conf
Oct 27 13:16:29 localhost NET[4059]:
/etc/sysconfig/network-scripts/ifup-post : updated /etc/resolv.conf
This entry was probably made right after the fresh installation.
I've done searches on the Internet for answers for this problem and the only thing I've found is that I am not the only one with this problem. But no body has offered any definite solution for it. I even posted the same question in the CentOS forum and haven't got any reply.
From what I have read, this problem may be caused by a bug or bugs in CentOS
5.x, but I hope it is not because I am in the process of switching from Fedora 10 to CentOS 5.4 and I don't want to get disappointed. Trivial as this may be, this does not give me confidence in CentOS distribution.
Can anyone come up with a solution?
Hi Rod,
Have you looked at this thread:
http://lists.centos.org/pipermail/centos/2009-October/083346.html
Maybe along the same lines as the problem that you are having not sure. HTH. Lee Perez
On Sat, Oct 31, 2009 at 7:37 PM, Lee Perez leecajun@windstream.net wrote:
Rod Rook wrote:
I installed CentOS 5.3 on Oct 27, 2009 and immediately updated it to 5.4.
I had not paid attention to this boot log until yesterday when I decided
to
take a look to do some trouble shooting.
Oct 27 13:16:24 localhost NET[3797]: /sbin/dhclient-script : updated
/etc/resolv.conf
Oct 27 13:16:29 localhost NET[4059]:
/etc/sysconfig/network-scripts/ifup-post : updated /etc/resolv.conf
This entry was probably made right after the fresh installation.
I've done searches on the Internet for answers for this problem and the
only
thing I've found is that I am not the only one with this problem. But no body has offered any definite solution for it. I even posted the
same
question in the CentOS forum and haven't got any reply.
From what I have read, this problem may be caused by a bug or bugs in
CentOS
5.x, but I hope it is not because I am in the process of switching from Fedora 10 to CentOS 5.4 and I don't want to get
disappointed.
Trivial as this may be, this does not give me confidence in CentOS distribution.
Can anyone come up with a solution?
Hi Rod,
Have you looked at this thread:
http://lists.centos.org/pipermail/centos/2009-October/083346.html
Maybe along the same lines as the problem that you are having not sure. HTH.
Hi, Lee, The problem mentioned in the thread you referred to is usually caused by Network Manager, which has nothing to do with boot.log. Thanks to you anyway.
Rod Rook wrote on Sat, 31 Oct 2009 23:55:52 -0500:
The problem mentioned in the thread you referred to is usually caused by Network Manager, which has nothing to do with boot.log.
You may want to elaborate what your *problem* really is. And please trim your answers, thanks.
Kai
Rod Rook wrote:
<snip>
Hi, Lee, The problem mentioned in the thread you referred to is usually caused by Network Manager, which has nothing to do with boot.log. Thanks to you anyway.
Perhaps he meant the thread beginning at http://lists.centos.org/pipermail/centos/2009-October/084836.html
My boot log looks like this:
[root@mavis log]# pwd /var/log [root@mavis log]# ls -l | grep boot -rw------- 1 root root 0 Nov 1 04:02 boot.log -rw------- 1 root root 0 Oct 25 04:02 boot.log.1 -rw------- 1 root root 0 Oct 18 04:02 boot.log.2 -rw------- 1 root root 0 Oct 11 04:02 boot.log.3 -rw------- 1 root root 0 Oct 4 04:02 boot.log.4 [root@mavis log]#
The zero length most like indicates that there are no entries :-; yet we can see that the machine has been rebooted:
[root@mavis log]# grep boot messages* messages.2:Oct 22 11:39:13 mavis shutdown[26207]: shutting down for system reboot messages.4:Oct 5 16:17:33 mavis shutdown[19739]: shutting down for system reboot messages.4:Oct 5 16:27:07 mavis shutdown[4979]: shutting down for system reboot messages.4:Oct 5 16:33:25 mavis shutdown[4738]: shutting down for system reboot messages.4:Oct 10 15:01:46 mavis shutdown[11680]: shutting down for system reboot [root@mavis log]#
Frankly, I can do very well without have boot.log at all but it seems that if we have it, it really should work! Mounting an old CentOS-4 disk, we see that it worked in CentOS 4:
[root@mavis log]# pwd /old-sys/var/log [root@mavis log]# ls -l | grep boot -rw------- 1 root root 107353 May 21 2007 boot.log -rw------- 1 root root 8867 May 20 2007 boot.log.1 -rw------- 1 root root 17265 May 13 2007 boot.log.2 -rw------- 1 root root 2224 May 6 2007 boot.log.3 -rw------- 1 root root 105 Apr 29 2007 boot.log.4 [root@mavis log]#
It's interesting to note that my current /etc/syslog.conf and the one in the older version both contain the entry:
# Save boot messages also to boot.log local7.* /var/log/boot.log
Beyond ferreting out those details, I am LOST.
On Sun, Nov 1, 2009 at 10:13 AM, Robert kerplop@sbcglobal.net wrote:
Rod Rook wrote:
<snip>
Hi, Lee, The problem mentioned in the thread you referred to is usually caused by Network Manager, which has nothing to do with boot.log. Thanks to you anyway.
Perhaps he meant the thread beginning at http://lists.centos.org/pipermail/centos/2009-October/084836.html
My boot log looks like this:
[root@mavis log]# pwd /var/log [root@mavis log]# ls -l | grep boot -rw------- 1 root root 0 Nov 1 04:02 boot.log -rw------- 1 root root 0 Oct 25 04:02 boot.log.1 -rw------- 1 root root 0 Oct 18 04:02 boot.log.2 -rw------- 1 root root 0 Oct 11 04:02 boot.log.3 -rw------- 1 root root 0 Oct 4 04:02 boot.log.4 [root@mavis log]#The zero length most like indicates that there are no entries :-; yet we can see that the machine has been rebooted:
[root@mavis log]# grep boot messages* messages.2:Oct 22 11:39:13 mavis shutdown[26207]: shutting down for system reboot messages.4:Oct 5 16:17:33 mavis shutdown[19739]: shutting down for system reboot messages.4:Oct 5 16:27:07 mavis shutdown[4979]: shutting down for system reboot messages.4:Oct 5 16:33:25 mavis shutdown[4738]: shutting down for system reboot messages.4:Oct 10 15:01:46 mavis shutdown[11680]: shutting down for system reboot [root@mavis log]#Frankly, I can do very well without have boot.log at all but it seems that if we have it, it really should work! Mounting an old CentOS-4 disk, we see that it worked in CentOS 4:
[root@mavis log]# pwd /old-sys/var/log [root@mavis log]# ls -l | grep boot -rw------- 1 root root 107353 May 21 2007 boot.log -rw------- 1 root root 8867 May 20 2007 boot.log.1 -rw------- 1 root root 17265 May 13 2007 boot.log.2 -rw------- 1 root root 2224 May 6 2007 boot.log.3 -rw------- 1 root root 105 Apr 29 2007 boot.log.4 [root@mavis log]#It's interesting to note that my current /etc/syslog.conf and the one in the older version both contain the entry:
# Save boot messages also to boot.log local7.* /var/log/boot.logBeyond ferreting out those details, I am LOST.
Robert, I responded to Lee's statement which referred to
http://lists.centos.org/pipermail/centos/2009-October/083346.html It is all there if you follow the thread. Anyway, you agree with me that there is a bug or bugs in CentOS 5.4. As I said earlier, it is not a critical matter, but it does not instill confidence in me about CentOS distro. What other bugs are there? The interesting thing, at least to me, is that there seems to be conspiratorial silence about such bugs. Nobody wants to speak ill of their beloved distro?
Rod Rook wrote:
Robert, I responded to Lee's statement which referred to http://lists.centos.org/pipermail/centos/2009-October/083346.html It is all there if you follow the thread. Anyway, you agree with me that there is a bug or bugs in CentOS 5.4. As I said earlier, it is not a critical matter, but it does not instill confidence in me about CentOS distro. What other bugs are there? The interesting thing, at least to me, is that there seems to be conspiratorial silence about such bugs. Nobody wants to speak ill of their beloved distro?
I'm sure there are other bugs. I'm equally sure there's no "conspiracy of silence" here, though. I guess the bug tracker would no longer exist if that was the case...
Robert wrote:
Rod Rook wrote:
Robert, I responded to Lee's statement which referred to http://lists.centos.org/pipermail/centos/2009-October/083346.html It is all there if you follow the thread. Anyway, you agree with me that there is a bug or bugs in CentOS 5.4. As I said earlier, it is not a critical matter, but it does not instill confidence in me about CentOS distro. What other bugs are there? The interesting thing, at least to me, is that there seems to be conspiratorial silence about such bugs. Nobody wants to speak ill of their beloved distro?
I'm sure there are other bugs. I'm equally sure there's no "conspiracy of silence" here, though. I guess the bug tracker would no longer exist if that was the case...
Going a bit further, the empty boot.log is listed in RedHat's bugzilla (https://bugzilla.redhat.com/buglist.cgi?quicksearch=boot.log) as bug # 529221 and 489838
The only reason I'm interested in it is that the nuances in syslog.conf still haven't penetrated my thick skull.
On Sun, Nov 1, 2009 at 11:31 AM, Robert kerplop@sbcglobal.net wrote:
Rod Rook wrote:
Robert, I responded to Lee's statement which referred to http://lists.centos.org/pipermail/centos/2009-October/083346.html It is all there if you follow the thread. Anyway, you agree with me that there is a bug or bugs in CentOS 5.4. As I said earlier, it is not a critical matter, but it does not instill confidence in me about CentOS distro. What other bugs are there? The interesting thing, at least to me, is that there seems to be conspiratorial silence about such bugs. Nobody wants to speak ill of their beloved distro?
I'm sure there are other bugs. I'm equally sure there's no "conspiracy of silence" here, though. I guess the bug tracker would no longer exist if that was the case...
Robert,
Thank you for having found some other sources to say it is a bug. Having used Fedora 8 and 10 for a long time and Fedora 11 for a few months and never experienced this in these distros, I can say this bug occurs only in old versions of Red Had distros like CentOS. Now that I know this is a bug, I will move on hoping that others will solve the problem
Rod Rook wrote:
Having used Fedora 8 and 10 for a long time and Fedora 11 for a few months and never experienced this in these distros, I can say this bug occurs only in old versions of Red Had distros like CentOS. Now that I know this is a bug, I will move on hoping that others will solve the problem
The problem with zero-length boot.log has existed in RHEL releases since 2007, and Red Hat has shown little interest in fixing it. You won't find an official fix from CentOS because CentOS strives to be bug-for- bug compatible with the corresponding RHEL release. Fedora, beginning with release 10, has made major changes to the boot sequence, and does write to boot.log. My guess is that you might see these changes show up in RHEL/CentOS 6. (RHEL 5 is based on Fedora Core 6.)
The data recorded in boot.log is typically not very interesting, so it's not likely you'll see a fix in RHEL 5. If you search on Google you can find some additions you can make in /etc/init.d/functions to record what init scripts were started. The only thing missing (from what I see in a Fedora 11 boot.log) would be the "OK" or "FAIL" status reports, but those can generally be inferred from what services started successfully.
2009/11/1 Rod Rook rod.rook@gmail.com:
Anyway, you agree with me that there is a bug or bugs in CentOS 5.4. As I said earlier, it is not a critical matter, but it does not instill confidence in me about CentOS distro. What other bugs are there? The interesting thing, at least to me, is that there seems to be conspiratorial silence about such bugs. Nobody wants to speak ill of their beloved distro?
Let's face it, it's pretty damned good for free!
Ben
-
Benjamin Donnachie -
Kai Schaetzl -
Lee Perez -
Robert -
Robert Nichols -
Rod Rook