Page 1 of 1
Update went wrong?!
Posted: Mon Dec 17, 2012 6:52 am
by mist_firefly
Pls help
Updated these 10 minutes ago and plesk is not responding anymore!!!
What can I do?
qmail-scanner x86_64 1:2.10-8.el5.art atomic 349 k
tortixd x86_64 2.2.23-8.el5.art asl-3.0 1.1 M
tortixd-mod_ssl
I also get on the control panel:
ERROR: PleskFatalException
Unable to connect to database: mysql_connect() [<a href='function.mysql-connect'>function.mysql-connect</a>]: Lost connection to MySQL server at 'waiting for initial communication packet', system error: 95
0: common_func.php3:150
psaerror(string 'Unable to connect to database: mysql_connect() [<a href='function.mysql-connect'>function.mysql-connect</a>]: Lost connection to MySQL server at 'waiting for initial communication packet', system error: 95')
1: auth.php3:107
ERROR: PleskFatalException
Unable to connect to database: mysql_connect() [<a href='function.mysql-connect'>function.mysql-connect</a>]: Lost connection to MySQL server at 'waiting for initial communication packet', system error: 95
0: common_func.php3:150
psaerror(string 'Unable to connect to database: mysql_connect() [<a href='function.mysql-connect'>function.mysql-connect</a>]: Lost connection to MySQL server at 'waiting for initial communication packet', system error: 95')
1: auth.php3:107
Re: Update went wrong?!
Posted: Mon Dec 17, 2012 7:19 am
by mist_firefly
No websites are working, no emails either
I can't login to the server through putty.
on our website we have:
Infinite loop detected in JError
Could it be the updates? or something else?
What can be done?
thanks
Re: Update went wrong?!
Posted: Mon Dec 17, 2012 7:59 am
by breun
Looks like at least your MySQL server isn't available.
What does service mysqld status say? If it's not running, can you start it using service mysqld start (psmon should do this automatically)? Any error messages in the MySQL server log file at /var/log/mysqld.log?
Other general things to check: Is the server under high load? Is there enough free diskspace?
Re: Update went wrong?!
Posted: Mon Dec 17, 2012 8:03 am
by mist_firefly
I am unable to login into the server at all.
We hope a restart will bring it back up but would I be able to rollback or downgrade the updates I made after?
What would be the best option?
Re: Update went wrong?!
Posted: Mon Dec 17, 2012 8:59 am
by mist_firefly
Server was manually restarted but it's still not responding I seem to have shell access..but very slow
It doesn't have rollback option in the files so I can't do that
cd /etc/yum.conf
vim /rpm/macros
Would a yum downgrade tortixd-mod_ssl
yum downgrade tortixd
do the trick?
Should I be worried about any dependencies?
What can I do? Nothing works
Re: Update went wrong?!
Posted: Mon Dec 17, 2012 9:04 am
by breun
I don't think the updates are directly causing your problem, but we don't know yet for sure. I'd start by determining what your problem actually is. Is the server under high load? If so, can you find out what process is causing high load?
Re: Update went wrong?!
Posted: Mon Dec 17, 2012 10:32 am
by mist_firefly
No shell access again.
Someone from our team is going to where the server is.
Could you please tell me some commands that would be helpful to finding out what caused the issue?
I was thinking about: top, ps -all, free -m
What else?
Thank you
Re: Update went wrong?!
Posted: Mon Dec 17, 2012 1:47 pm
by mist_firefly
server was turned off and back on from the button
it started all the processes:
ip info eth1 failed
psa firewall:disabled failed
but after a while screen was flashing and showing: call trace .... [and a lot of information] : blocked
and multiple lines with that for different processes
What does that mean?
How can I login to it? or get to the command line to check logs and see what is happening?
thanks
Re: Update went wrong?!
Posted: Mon Dec 17, 2012 2:32 pm
by mist_firefly
On the screen at the server:
tortixd:apr_sockaddr_info_get() failed for plesk.domain.xxx
tortixd: could not reliably determine the server's fully qualified doamin name, using 127.0.0.1 for server name
failed to start
httpd failed to start
named service failed to start
mysqld failed also
Could anybody help with some advice on what can be done in these situations?
Re: Update went wrong?!
Posted: Mon Dec 17, 2012 3:26 pm
by mist_firefly
update:
managed to boot as single user
ssh working slowly but to keep it running i need to keep killing qmail processes ..as that slows everything down and makes it stop
what could i do ?
should i down grade?
or?
Re: Update went wrong?!
Posted: Mon Dec 17, 2012 5:13 pm
by mikeshinn
ssh working slowly but to keep it running i need to keep killing qmail processes ..as that slows everything down and makes it stop
What are the qmail processes doing? Given that you changed qmailscanner, its possible you may need to downgrade perhaps its not compatible with your configuration. I'd find out what qmail is doing first, this could all be caused by something else too.
Also, these:
tortixd x86_64 2.2.23-8.el5.art asl-3.0 1.1 M
tortixd-mod_ssl
Have no effect on your system, they are only used by ASL. They would not cause anything you have reported.
Re: Update went wrong?!
Posted: Mon Dec 17, 2012 5:34 pm
by mist_firefly
Yes qmail-scanner seems to be the issue I have a post open here
https://www.atomicorp.com/forums/viewto ... f=4&t=6463 with more details on what we have.
doing top results in a lot of processes like:
23894 qscand 20 0 50500 8732 2344 S 2.6 0.1 0:00.08 perl5.8.8
23908 qscand 20 0 50532 8752 2344 S 2.6 0.1 0:00.08 perl5.8.8
which are slowing down the server
What would be best to be done as now emails are not working
Websites are but only because of the cron job that kills all qmail processes that keep starting
Re: Update went wrong?!
Posted: Mon Dec 17, 2012 6:47 pm
by breun
Re: Update went wrong?!
Posted: Mon Dec 17, 2012 11:00 pm
by coolemail
Thanks breun,
tried doing what you proposed, and initially it looked OK:
removed qmail-scanner
Installed Postfix (or I think so). No evidence of it on the server which still shows things like below:
Dec 18 02:57:11 plesk3 X-Qmail-Scanner-2.08st: [plesk3.emailitis.co.uk135579943179825246] clamdscan: corrupt or unknown clamd scanner error or memory/resource/perms problem - exit status 512/2
Dec 18 02:57:11 plesk3 spamdyke[25188]: DENIED_OTHER from:
20121218000506ed8a5344bfea41dcaac803e23657b16e-c2xmghcc95oucg@bounces.amazon.com to:
angela@hosted-domain.co.uk origin_ip: 193.111.201.135 origin_rdns: lon-mail-relay-2.mail-relay.net auth: (unknown) encryption: TLS reason: 451_qq_temporary_problem_(#4.3.0)
Mike, Does that give enough to see what might be happening?
[root@plesk3 ~]# /usr/local/psa/admin/sbin/mailmng --features | grep SMTP_Server
$features['SMTP_Server'] = "Postfix";
$features['SMTP_Server_package'] = "postfix";
[root@plesk3 ~]#
but postfix appears not to be working:
[root@plesk3 ~]# service postfix status
master is stopped
[root@plesk3 ~]# service postfix start
Starting postfix: [ OK ]
[root@plesk3 ~]# service postfix status
master is stopped
[root@plesk3 ~]#
and doing
Run /usr/local/psa/admin/bin/autoinstaller again and switch back to Qmail as per the post is not going back to qmail - it must be me but not sure what I have (not) done.
and it looks like clamdscan is getting in the way somewhere from the maillog entry there.
Trying to reinstall the qmail-scanner again hung the server:
[root@plesk3 ~]# yum install qmail-scanner
Loaded plugins: fastestmirror
Loading mirror speeds from cached hostfile
* asl-3.0: www6.atomicorp.com
* atomic: www7.atomicorp.com
* base: mirror01.th.ifl.net
* extras: mirror01.th.ifl.net
* rpmforge:
http://www.mirrorservice.org
* updates: mirror01.th.ifl.net
Excluding Packages in global exclude list
Finished
Setting up Install Process
Resolving Dependencies
--> Running transaction check
---> Package qmail-scanner.x86_64 1:2.10-8.el5.art set to be updated
--> Finished Dependency Resolution
Dependencies Resolved
=============================================================================================================================================
Package Arch Version Repository Size
=============================================================================================================================================
Installing:
qmail-scanner x86_64 1:2.10-8.el5.art atomic 349 k
Transaction Summary
=============================================================================================================================================
Install 1 Package(s)
Upgrade 0 Package(s)
Total download size: 349 k
Is this ok [y/N]: y
Downloading Packages:
qmail-scanner-2.10-8.el5.art.x86_64.rpm | 349 kB 00:00
Running rpm_check_debug
Running Transaction Test
Finished Transaction Test
Transaction Test Succeeded
Running Transaction
Creating mailbox file: File exists
useradd: warning: the home directory already exists.
Not copying any file from skel directory into it.
Installing : qmail-scanner 1/1
Installed:
qmail-scanner.x86_64 1:2.10-8.el5.art
Complete!
[root@plesk3 ~]# cp /etc/qmail-scanner.ini /etc/qmail-scanner-kuhle-ini-backup-181212.bak
[root@plesk3 ~]# mv /etc/qmail-sacnner.ini.rpmsave /etc/qmail-scanner.ini
mv: cannot stat `/etc/qmail-sacnner.ini.rpmsave': No such file or directory
[root@plesk3 ~]# mv /etc/qmail-scanner.ini.rpmsave /etc/qmail-scanner.ini
mv: overwrite `/etc/qmail-scanner.ini'? y
[root@plesk3 ~]# qmail-scanner-reconfigure
Reconfiguring spamassassin settings: Done
Reconfiguring clamav settings: Done
Configuring qmail-scanner: