• 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

How to Update from 7.1.7 to 7.5.4+ [without updater]

etc/hosts:
Code:
 #Do not remove the following line, or various programs
# that require network functionality will fail.
127.0.0.1               localhost.localdomain localhost
64.xx.xx.123            websrv."mydomain".com websrv
without the xx's and "mydomain" of course

The ip I usually ssh into and access with plesk is 64.xx.xx.124

etc/resolv.conf
Code:
search chi.us.siteprotect.com
nameserver 64.26.0.152
nameserver 64.26.0.151
But for as long as I can remember, whenever I ssh into the server it says
Code:
Address 64.41.74.124 maps to ltf209.chi.us.siteprotect.com, but this does not map back to the address - POSSIBLE BREAKIN ATTEMPT!
I didn't set it up originally and It was never a problem... mail worked fine
Also, I can ping mail."mydomain".com fine even with this message.

I haven't changed any of these files or the DNS or IP settings (knowingly) recently.
 
Here is an example of one of the domain's DNS settings under Plesk > Domains > "mydomain" > Services > DNS

Code:
[url]www.[/url]"mydomain".com.	CNAME	"mydomain".com.	
webmail."mydomain".com.	A	64.xx.xx.124	
mssql."mydomain".com.	A	64.xx.xx.124	
mail."mydomain".com.	A	64.xx.xx.124	
"mydomain".com.		NS	ns.anotherdomain.com.	
"mydomain".com.		NS	ns2.anotherdomain.com.	
"mydomain".com.		A	64.xx.xx.124	
"mydomain".com.		MX (10)	mail."mydomain".com.	
ftp."mydomain".com.	CNAME	"mydomain".com.	
64.xx.xx.124 / 24	PTR	"mydomain".com.	
10 DNS records total
 
Remove the quotation marks in every instance of the domain name - totally - in every file and the dns settings!
 
the quatation marks really aren't there :) and instead of mydomain it lists the actual domain

In the mail preferences, POP3 Authentication was checked and STMP was unchecked... so I reversed it... clicked ok, tried to send an email... same problem though...
 
Qmail seems to be working hard... seems to be an entry in the log every second at least... all spam I think... there are probably on 10-15 mail accounts on the entire server
Code:
Apr 11 08:59:58 websrv qmail: 1144763998.551020 starting delivery 105: msg 8011938 to remote [email][email protected][/email]
Apr 11 08:59:58 websrv qmail: 1144763998.551039 status: local 0/10 remote 20/20
Apr 11 09:00:00 websrv qmail: 1144764000.890828 delivery 79: deferral: Sorry,_I_wasn't_able_to_establish_an_SMTP_connection._(#4.4.1)/
Apr 11 09:00:00 websrv qmail: 1144764000.890900 status: local 0/10 remote 19/20
Apr 11 09:00:00 websrv qmail: 1144764000.890949 starting delivery 106: msg 8013182 to remote [email][email protected][/email]
Apr 11 09:00:00 websrv qmail: 1144764000.890973 status: local 0/10 remote 20/20
Apr 11 09:00:01 websrv qmail: 1144764001.071996 delivery 106: deferral: Sorry,_I_wasn't_able_to_establish_an_SMTP_connection._(#4.4.1)/
Apr 11 09:00:01 websrv qmail: 1144764001.072037 status: local 0/10 remote 19/20
Apr 11 09:00:01 websrv qmail: 1144764001.072109 starting delivery 107: msg 8012409 to remote [email][email protected][/email]
Apr 11 09:00:01 websrv qmail: 1144764001.072125 status: local 0/10 remote 20/20
 
ok seems someone else had a similar problem:here
They eventually solved it, something about "xinted expected a file call smtp_psa while smtp.psa in the dir"

um, so I don't know what xinted is or how I can check the smtp_psa or fix it... the author of the thread also encountered this problem when upgrading from 7.1.x to 7.5.4 where there was some sort of bug in the upgrade package

Could this be my problem too?
Code:
[root@websrv]# locate smtp_psa
/etc/xinetd.d/smtp_psa
[root@websrv]# locate smtp.psa
[root@websrv]#
which seems to indicate that the smtp_psa file is named correctly.

Code:
#telnet localhost 25 
Trying 127.0.0.1...
Connected to localhost.localdomain (127.0.0.1).
Escape character is '^]'.
relaylock: Error: HTTPD_INCLUDE_D not defined
relaylock: Error: HTTPD_INCLUDE_D not defined
Connection closed by foreign host.

#telnet localhost 110
Trying 127.0.0.1...
Connected to localhost.localdomain (127.0.0.1).
Escape character is '^]'.
+OK Hello there. <[email protected]>

and after using the autoupdater, I think some of my psa versions might be conficting with the email program:
Code:
# rpm -qa | grep psa
frontpage-5.0-51psa.rhel3
perl-FreezeThaw-0.43-44psa
psa-api-8.0.0-rhel3.build80060404.15
perl-MLDBM-2.01-44psa
perl-HTML-Format-1.23-44psa
psa-pylibplesk-8.0.0-rhel3.build80060331.13
psa-backup-manager-8.0.0-rhel3.build80060404.15
perl-HTML-Tree-3.16-44psa
psa-boost-1.32.0-6.rhel3
psa-migration-manager-8.0.0-rhel3.build80060404.15
psa-locale-base-en-7.1.7-rhel3.build71050228.12
perl-MLDBM-Sync-0.30-44psa
psa-hotfix5-7.5.4-rhel3.build75060118.18
psa-logrotate-3.7-rhel3.build80060331.13
psa-autoinstaller-3.0.1-build060406.10
psa-agent-1.4.0-80060331.13
psa-pre-keyupdate-7.5.4-rhel3.build75050824.12
psa-hotfix6-7.5.4-rhel3.build75060321.20
psa-qmail-rblsmtpd-0.70-rhel3.build71040817.01
perl-Font-AFM-1.18-44psa
psa-7.5.4-rhel3.build75050824.12
psa-horde-3.0.5-rhel3.build75050824.12
psa-qmail-1.03-rhel3.build80060331.13
psa-proftpd-xinetd-1.2.10-rhel3.build80060331.13
psa-tomcat-configurator-8.0.0-rhel3.build80060331.13
psa-libpam-plesk-8.0.0-rhel3.build80060331.13
psacct-6.3.2-27
psa-key-7.5-rhel3.build75050824.12
psa-courier-imap-add-8.0.0-rhel3.build80060331.13
psa-api-rpc-8.0.0-rhel3.build80060404.15
psa-imp-4.0.3-rhel3.build75050824.12
psa-proftpd-1.2.10-rhel3.build80060331.13
psa-php4-configurator-1.0.0-rhel3.build80060329.02
psa-locale-base-en-US-8.0.0-rhel3.build80060331.13
mod_jk-ap20-1.2.15-42psa.rhel3
which is similar to someone in this thread about 7.54 to 8.0.0 problems
 
Mail Problems

At this point are you able to send mail through webmail? Or are you not able to log in to Webmail at all?

If you can log in to webmail, any mail that was supposed to go to that account is not arriving, and this is for all accounts? Try installing qmHandle (http://sourceforge.net/projects/qmhandle) if you don't have it yet, and run qmHandle -s to see if there's anything stuck in the queue.

Qmail is running along with courier-imap right? Just to be sure, try service courier-imap restart
and
service qmail restart

I'm just trying to get an idea of whether it's incoming mail or outgoing mail that's the problem, because if it's only incoming them smtp_psa shouldn't matter at all.

Jordan
 
No I can't log into webmail, it says login failed everytime.
Both services are running and just to be sure I restarted both.
Based on the thread I linked to above, I uninstalled psa-spamassasin

I have no idea now... can't send or recieve... should I try to upgrade to 8.0.0 for all psa's? Maybe that will help since I have some 8.0.0 rpms, some 7.5.4 rpms and some 7.1.7 rpms
 
Question, when you showed your /etc/hosts file you show the IP address as ending with .123 but your dns settings end in .124. There should also be an "A" record for websrv.mydomain.com.

Also, the "search" line in /etc/resolve.conf is not necessary, not that it is causing the problem, just FYI
 
I just checked on that
xx.xx.xx.123 is a name server with Hostway
xx.xx.xx.124 is my dedicated webserver purchased through Hostway

I am really terrible at understanding DNS and ips but the .123 address would then point to .124 which is websrv.mydomain.com

Should I change this? I would MUCH rather have the mail down than the website down... but if its safe and correct then I'll make the change.


As for the mail settings... I just tried to add a mail account through Plesk and it just spins and spins and never successfully adds the account....

this thread thought the problem could also have to do with Dr.Web so I stopped the drwebd service.
I'm not noticing anything different though.
 
Definitely change your /etc/hosts file to look like this:

#Do not remove the following line, or various programs
# that require network functionality will fail.
127.0.0.1 localhost.localdomain localhost
64.xx.xx.124 websrv.mydomain.com websrv

Then add an "A" record to the domain's DNS like this:

websrv.mydomain.com A 64.xx.xx.124

Then restart named from the control panel or through SSH like:

# /etc/init.d/named restart
 
I made those changes but another domain on the server has this for DNS Settings:
Keep in mind that .123 is the Hostway nameserver and .124 is the actual dedicated server

I changed this actual name of this second domain to SECOND-DOMAIN
Code:
Restoring the DNS zone by the template
IP Address: 	64.xx.xx.123
WWW prefix: 	Yes

[url]www.SECOND-DOMAIN.com.	CNAME	SECOND-DOMAIN.com.	[/url]
websrv.SECOND-DOMAIN.com.	A	64.xx.xx.123	
webmail.SECOND-DOMAIN.com.	A	64.xx.xx.123	
uk.SECOND-DOMAIN.com.	A	64.xx.xx.123	
mssql.SECOND-DOMAIN.com.	A	64.xx.xx.123	
mail.SECOND-DOMAIN.com.	A	64.xx.xx.123	
ftp.SECOND-DOMAIN.com.	CNAME	SECOND-DOMAIN.com.	
de.SECOND-DOMAIN.com.	A	64.xx.xx.123	
db2.SECOND-DOMAIN.com.	A	64.xx.xx.125	
db1.SECOND-DOMAIN.com.	A	64.xx.xx.126	
64.xx.xx.123 / 24	PTR	SECOND-DOMAIN.com.	
SECOND-DOMAIN.com.		NS	ns.websiteinternethosting.com.	
SECOND-DOMAIN.com.		NS	ns2.websiteinternethosting.com.	
SECOND-DOMAIN.com.		A	64.xx.xx.123	
SECOND-DOMAIN.com.	MX (10)	mail.SECOND-DOMAIN.com.	
15 DNS records total
Does this look ok? It works alright :)

Here is the tail of the maillog after making those DNS changes on the first domain:
Code:
Apr 11 11:35:22 websrv qmail: 1144773322.941923 starting delivery 6: msg 8011948 to remote [email][email protected][/email]
Apr 11 11:35:22 websrv qmail: 1144773322.941974 status: local 0/10 remote 2/20
Apr 11 11:35:22 websrv qmail: 1144773322.949540 delivery 6: failure: 
Sorry._Although_I'm_listed_as_a_best-preference_MX_or_A_for_that_host,/it_isn't_in_my_control/locals_file,_so_I_don't_treat_it_as_local._(#5.4.6)/
Apr 11 11:35:22 websrv qmail: 1144773322.949754 status: local 0/10 remote 1/20
Apr 11 11:35:22 websrv qmail: 1144773322.949787 triple bounce: discarding bounce/8011948
Apr 11 11:35:22 websrv qmail: 1144773322.949827 end msg 8011948
Apr 11 11:35:32 websrv qmail: 1144773332.948548 starting delivery 7: msg 8012978 to remote [email][email protected][/email]
Apr 11 11:35:32 websrv qmail: 1144773332.948615 status: local 0/10 remote 2/20
Apr 11 11:35:36 websrv qmail: 1144773336.298896 delivery 7: deferral: Sorry,_I_wasn't_able_to_establish_an_SMTP_connection._(#4.4.1)/
Apr 11 11:35:36 websrv qmail: 1144773336.298938 status: local 0/10 remote 1/20
and I think this is a user trying to access their email (?)
Code:
Apr 11 11:51:53 websrv pop3d: Connection, ip=[71.xxx.xxx.23]
Apr 11 11:51:53 websrv pop3d: IMAP connect from @ [71.xxx.xxx.23]INFO: LOGIN, user=heather, ip=[71.xxx.xxx.23]
Apr 11 11:51:53 websrv pop3d: 1144774313.852478 LOGOUT, user=heather, ip=[71.xxx.xxx.23], top=0, retr=0, time=0, rcvd=12, 
sent=39, maildir=/var/qmail/mailnames/SECOND-DOMAIN.com/heather/Maildir
 
Man, that looks messed up. If nothing else, you have this domain's PTR (reverse DNS) record pointing to a nameserver! Plus, you have the mail record pointing to a nameserver. Your DNS is screwed.

If I ping ns.websiteinternethosting.com I get the 64.xx.xx.123 IP. But if I ping ns2.websiteinternethosting.com I get the IP for your dedicated server as to be pinged but it pings a totally different IP. Which happens to be Hostway's.

I'd contact your host and try to get them to help you straighten this out.
 
Although I wasn't involved in setting the servers up when they were first setup, I overheard a conversation about wanting a different nameserver that was on a different ip....
so I think it was requested to be like that (?)
I've known that the DNS settings were always a little screwy because I have to turn off the iptables service on reboots.

I'm not sure if this is the cause of the problem though..
 
Go to dnsstuff.com and check these IP's out. The 124 IP has a reverse DNS pointing to brothers.brothersreunited.com. I don't see that anywhere in your dns settings.
 
I think the DNS report is ok, I didn't see any brothersunited references for the ip I'm using

I just checked some DNS info and found that there are 2 nameservers:
xx.xx.xx.123 = NS1.adomain.com [nameserver only]
xx.xx.xx.124 = NS2.adomain.com [Dedicated - Plesk is installed on this as well as all domains, files etc]

I've been told that "the ns. and ns2. are set up with the domain registrer so they point to 123 and 124"

So does this still seem "messed up"?
Should I change anything?

Would it change anything if I said I really don't care about Plesk at all? :) Its nice to have sometimes but I'd rather have PHP5. Can I just do away with all these mismatching rpm packages and completely reinstall Plesk or is that even more difficult?
 
I just tried to run the autoinstaller via the command line and tried to just install what was recommended in 7.5.4 and got this error
Code:
Reading system installed packages...done.
Download file plesk-7.5.4-rhel3-i386.hdr.gz: 40%..100% done
Download file plesk-7.5.4-rhel3-i386.hdr.gz: 100% done
Resolve components
ERROR: Can't add to black list product package psa-locale-base-en-US
ERROR: Installation failed
I think it might be because I have some 8.0.0 rpms... should I get rid of them... if so, how (rpm -e <package>???)

or is it possible to just force install for 7.5.4 or 8.0.0... as long as it won't break the website...
How would be the best way to do this?
 
So is removing all psa packages and downloading the rpm for 8.0.0 and forcing the install a good idea?
 
Back
Top