• If you are still using CentOS 7.9, it's time to convert to Alma 8 with the free centos2alma tool by Plesk or Plesk Migrator. Please let us know your experiences or concerns in this thread:
    CentOS2Alma discussion

7.5.1 -> 7.5.2 Server will not boot now

C

cYon

Guest
I just tried to upgrade my server from 7.5.1 to 7.5.2 and I received the following error message:
Trying to stop MySQL server... done
There is no /etc/localtime file in system.
Run tzsetup as root to set up your timezone
error: %pre(psa-7.5.2-fc2.build75050128.11) scriptlet failed, exit
status 1
error: install: %pre scriptlet failed (2), skipping
psa-7.5.2-fc2.build75050128.11
===> Reading /etc/psa/psa.conf ...
===> Checking database
ERROR 2002: Can't connect to local MySQL server through socket
'/var/lib/mysql/mysql.sock' (2)
===> Reading /etc/psa/psa.conf ...
===> Setting up file owners
===> Installing database
ERROR 2002: Can't connect to local MySQL server through socket
'/var/lib/mysql/mysql.sock' (2)
ERROR 2002: Can't connect to local MySQL server through socket
'/var/lib/mysql/mysql.sock' (2)
error: %post(cs-gs-1.0.0-fc2.build75050128.11) scriptlet failed, exit
status 1
===> Reading /etc/psa/psa.conf ...
===> Setting up file owners
===> Installing database
ERROR 2002: Can't connect to local MySQL server through socket
'/var/lib/mysql/mysql.sock' (2)
ERROR 2002: Can't connect to local MySQL server through socket
'/var/lib/mysql/mysql.sock' (2)
error: %post(psa-firewall-1.0.1-fc2.build75050128.11) scriptlet failed,
exit status 1
===> Reading /etc/psa/psa.conf ...
/usr/local/psa/admin/bin/httpsdctl stop: httpd stopped
Stopping Plesk: [ OK ]
===> Reading /etc/psa/psa.conf ...

----- end of output -----


Now when I go to Updater it says the base plesk files are not installed and when I try to install them I get the following error:
There is no /etc/localtime file in system.
Run tzsetup as root to set up your timezone
error: %pre(psa-7.5.2-fc2.build75050128.11) scriptlet failed, exit
status 1
error: install: %pre scriptlet failed (2), skipping
psa-7.5.2-fc2.build75050128.11
----- end of output -----

I rebooted my server and now it fails to boot. I am able to gain access via rescure mode, however, I have no clue on what's keeping it from booting. I checked boot.log and dmesg and found no errors. Does anyone have any ideas?
 
It told you each run to set your time zone.. if you would listen to it :rolleyes:
 
Sadly, I did not visit the forums before upgrading, a lesson learned the hard way
 
I had to just start over and format the server. I couldn't find a fix fast enough, and had to get the server back online.
 
Ouch.
In my logs it said eth0 unexpected mac address
bringing up eth0 failed

I guess somehow the mac changed or the mac in the startup script did.
After commenting out the address it worked.
Mine was 1&1 also, now I have a different problem that will need a new post.

Thanks for the reply though
 
Originally posted by crshjnke
Ouch.
In my logs it said eth0 unexpected mac address
bringing up eth0 failed

I guess somehow the mac changed or the mac in the startup script did.
After commenting out the address it worked.
Mine was 1&1 also, now I have a different problem that will need a new post.

Thanks for the reply though

I had the same thing happen on my 1and1 server (FC2). They had to restore the network card configuration to DHCP, then my system was at least hobbling along, though Plesk, PHP and some other things were broken.
 
Originally posted by wchatham
I had the same thing happen on my 1and1 server (FC2). They had to restore the network card configuration to DHCP, then my system was at least hobbling along, though Plesk, PHP and some other things were broken.

Snap. 1&1 (UK) with FC2. Took the support team 30 minutes to figure out this problem. I had already diagnosed the problem, just wanted them to point out what needed to be changed.

Only problem I have noticed after the server was accessible again was that httpsdctl had to be manually restarted.
 
Wythrol,

I had the same problem upgrading to 7.5.2 (FC2) with 1and1. Their Tech Support could not provide me with an answer after almost 2 hours of talking to them.

Here is the fix I came up with on my own .

I went in through the recovery console and noticed in the boot log the message that the MAC address had changed and that eth0 was not starting. It occured to me that eth0 must be up and running when I'm in the recovery console so... I ran ifconfig and looked at the HWADDR (MAC) that was being used for eth0 in recovery console.

Then I edited the file /etc/sysconfig/network-scripts/ifcfg-eth0
and changed the HWADDR line to use the MAC address that was being used in the recovery console and restarted the machine. It came up perfectly.

Now I don't pretend to know what happened in the first place but it seemed to me that if the MAC address had changed then I should try to supply it with the coreect (or new?) MAC address.

Hope this helps other 1and1 people with similar problems.


P.S. I told 1and1 tech support how to do this fix... doesn't mean they will use it...too bad.
 
What I would like to know is how it got changed?
Did psa update somehow affect this?

Never had a problem rebooting until the 7.5.2 update.
Atleast it boots now I just need to fix my other problems.
 
I don't know... Plesk says it wasn't their problem... but I HAVE had this problem once before after an upgrade.... I'm not sure if the root cause of the problem was the same but 1and1 fixed it in about an hour and a half...said that some file had become corrupted and they "just rebuilt it"

I'm pretty certain that it was the same problem though...that's why I was so ticked that 1and1 couldn't immediately provide me with an answer to a problem they had fixed once before.

Did the fix I gave provide some relief for you or had you already done that to get your server back up?
 
ZopfWare,

When my server failed to come back online after installing the Base Packages update and rebooting I dropped down into the Rescue System, checked the logs and spotted immediately that the MAC addressing was causing a problem. However, not knowing linux as well as I should, finding the config file that needed changing was causing me problems.

I only resort to trying to deal with 1&1 tech support when forced to, and I generally try and call late in the evening when the UK operations close and calls are transfered to the US whom I find marginally less frustrating to deal with than their UK collegues.

However, on this occasion UK server support were still open and I was subjected to a 30 minute call while the support chap hmm'd and aahhh'd and sucked air through his teeth before concluding that the MAC address was wrong and that if I edited /etc/sysconfig/network-scripts/ifcfg-eth0
things would work out.. Considering 1&1 request a customised version of PSA from Plesk you would think that they would be aware of these issues before the updates are released.
 
Originally posted by ZopfWare
I had the same problem upgrading to 7.5.2 (FC2) with 1and1.
...

Here is the fix I came up with on my own .

I edited the file /etc/sysconfig/network-scripts/ifcfg-eth0
and changed the HWADDR line to use the MAC address that was being used in the recovery console and restarted the machine. It came up perfectly.

Hope this helps other 1and1 people with similar problems.

Thanks for posting this ZopfWare, I had exactly the same problem and this was the solution! I just wish 1and1 had a forum/kb where stuff like this was available!
 
Originally posted by alanjshea
Thanks for posting this ZopfWare, I had exactly the same problem and this was the solution! I just wish 1and1 had a forum/kb where stuff like this was available!

A chap from here set up an unofficial forum for 1&1 users over at http://www.1and1faq.com
it may be one to add to your bookmarks.
 
god i hate 1and1 support. I had this issue, and was unable to identify the issue prior to visiting this thread. I used the suggestions in this thread to solve the problem.

1and1 support told me "you purchased a root server, so you are on your own"

*%^*% you 1and1. This is a plesk problem and plesk was included on my server when it was initially imaged. I beleive that any plesk support does in fact fall on 1and1's hands. Especially for f'ed up updates like 7.5.1-7.5.2 has been.

In closing. F you 1and1, please learn to support your customers better, especially with a widespread problem like this. Dont commit your paying customers to a brick wall like "you purchased a root server, so you are on your own" </rant>
 
Back
Top