Koozali.org: home of the SME Server

where did the backup configuration page go?

Offline julianop

  • *
  • 61
  • +0/-0
where did the backup configuration page go?
« on: July 08, 2023, 08:00:37 PM »
I'm running koozali 10, replacing 9 and previous versions

I just noticed that I don't have a backup configuration page in Configuration Manager. I had recently installed SMEadmin, and the meu item for that shows up in the Administration group where "Backup" used to be in in version 9.

How do I configure backups now?

Offline TerryF

  • grumpy old man
  • *
  • 1,819
  • +6/-0
Re: where did the backup configuration page go?
« Reply #1 on: July 09, 2023, 12:22:02 AM »
A clean VM, Backup or Restore item is present in Server-Manager, install Smeserver-smeadmin, check Server-Manager, Backup or Restore item is present as second item below new Sysadmin item, so something else you may have been doing perhaps?
--
qui scribit bis legit

Offline mmccarn

  • *
  • 2,625
  • +10/-0
Re: where did the backup configuration page go?
« Reply #2 on: July 09, 2023, 01:27:08 PM »
On my SME10 (v10.1, up-to-date) with no extra backup contribs installed, "Backup or restore" sits immediately under "Administration"

Offline julianop

  • *
  • 61
  • +0/-0
Re: where did the backup configuration page go?
« Reply #3 on: July 09, 2023, 06:59:58 PM »
No, it's no longer there; that's my point.
I realize that it should be there, but it's not. And no, I haven't been doing "something else" :-)

It was there, but now it ain't. And the weird thing is that I'm sure, when I replaced my 9.1 install with 10, I reconfigured backup, but when I look at my external drive there is nothing in the directory. It sure seems that I missed backup config altogether on this new 10 install... how embarrassing - I really am getting senile!

If I could figure out how to insert a clip here, I'd do that, but I think my text explanation is clear enough: the "backup" menu item is now missing, having been replaced by the "Smeadmin" item.

Should I uninstall SMEadmin and see if the backup menu item shows up again??
« Last Edit: July 09, 2023, 07:39:49 PM by julianop »

Offline ReetP

  • *
  • 3,704
  • +5/-0
Re: where did the backup configuration page go?
« Reply #4 on: July 09, 2023, 08:02:11 PM »
You should not need to insert anything. It should just work. Lets find out what went went wrong and where and try to fix it properly rather than ending up hacking our way into more trouble and creating an https://xyproblem.info/ :-)

Follow this vaguely please:

https://forums.koozali.org/index.php/topic,54724.0.html

So first please give us a little bit of history. Describe how you did your upgrade? Did you notice any errors?

Please run these and post the output:

Code: [Select]
/sbin/e-smith/audittools/newrpms
/sbin/e-smith/audittools/templates

Also have a look in that post above at how to search your logs for errors and see if you can see anything obvious.

...
1. Read the Manual
2. Read the Wiki
3. Don't ask for support on Unsupported versions of software
4. I have a job, wife, and kids and do this in my spare time. If you want something fixed, please help.

Bugs are easier than you think: http://wiki.contribs.org/Bugzilla_Help

If you love SME and don't want to lose it, join in: http://wiki.contribs.org/Koozali_Foundation

Offline julianop

  • *
  • 61
  • +0/-0
Re: where did the backup configuration page go?
« Reply #5 on: July 09, 2023, 10:05:09 PM »
Hi (again) ReetP, and thanks for helping. I'm still trying to get around to two other issues, but having no backups really demands immediate attention.

Item 1:
Code: [Select]
[root@dejavu log]# /sbin/e-smith/audittools/newrpms
Loaded plugins: fastestmirror, post-transaction-actions, priorities, smeserver
Loading mirror speeds from cached hostfile
 * base: mirror.steadfastnet.com
 * smeaddons: mirror.canada.pialasse.com
 * smeos: mirror.canada.pialasse.com
 * smeupdates: mirror.canada.pialasse.com
 * updates: tx-mirror.tier.net
Extra Packages
GeoIP.x86_64                            1.6.12-9.el7.sme            @smecontribs
GeoIP-GeoLite-data.noarch               2018.06-7.el7.sme           @smecontribs
GeoIP-GeoLite-data-extra.noarch         2018.06-7.el7.sme           @smecontribs
hddtemp.x86_64                          0.3-0.31.beta15.el7         @smecontribs
smeserver-certificate.noarch            0.0.4-13.el7.sme            @smecontribs
smeserver-fetchmail.noarch              1.6-26.el7.sme              @smecontribs
smeserver-smeadmin.noarch               1.6-10.el7.sme              @smecontribs
[root@dejavu log]#

Item 2
Code: [Select]
[root@dejavu log]# /sbin/e-smith/audittools/templates
/etc/e-smith/templates-custom/etc/e-smith/pam/accounts.deny/10passwordRequired: MANUALLY_ADDED, OVERRIDE
[root@dejavu log]#

To the other questions, this "upgrade" was a fresh install onto new hardware, per the wiki.... obviously, because 10 isn't an in-place install from 9.1. It was performed, IIRC, at the end of May last year.
I didn't notice any errors, and the system came up quite easily.
The only two problems I'm aware are those for which I already have posts on, and am trying to get around to investigating:
1. The persistent "Warning: a reconfigure and reboot is required before proceeding!" message that won't go away; and
2. The problem I had running the "Manage SSL Certificates" add-in, which I had to circumvent by installing a updated SSL certificate manually, and which gave me the "Error: CSRF token is invalid or outdated." error when I tried to use the add-in.
Other than that, the system is working as I expect/want it to.

I should note that other than the "certificate", "fetchmail", and "smeadmin" features, I was not consciously responsible for any of the other modifications. This machine functions as a mail server and a file server for my family and me, and does nothing else.

Continuing with following recommendations, I se this:
Code: [Select]
[root@dejavu log]# less /var/log/messages | grep -iE "uninitialized|WARNING|ERROR"
Jul  9 04:35:22 dejavu smartd[785]: Device: /dev/sdc [SAT], SMART Usage Attribute: 7 Seek_Error_Rate changed from 200 to 100
Jul  9 07:05:22 dejavu smartd[785]: Device: /dev/sdb [SAT], SMART Usage Attribute: 7 Seek_Error_Rate changed from 100 to 200
Jul  9 07:43:17 dejavu kernel: [797848.791541] cvm-unix[26942]: segfault at 0 ip 00007f92add35037 sp 00007ffda3a8b5a8 error 4 in libc-2.17.so[7f92adbf6000+1c4000]
Jul  9 07:43:17 dejavu kernel: cvm-unix[26942]: segfault at 0 ip 00007f92add35037 sp 00007ffda3a8b5a8 error 4 in libc-2.17.so[7f92adbf6000+1c4000]
Jul  9 08:05:22 dejavu smartd[785]: Device: /dev/sdc [SAT], SMART Usage Attribute: 7 Seek_Error_Rate changed from 100 to 200
Jul  9 11:56:20 dejavu kernel: [813031.358964] cvm-unix[2979]: segfault at 0 ip 00007fd86d0ad037 sp 00007ffc6046f718 error 4 in libc-2.17.so[7fd86cf6e000+1c4000]
Jul  9 11:56:20 dejavu kernel: cvm-unix[2979]: segfault at 0 ip 00007fd86d0ad037 sp 00007ffc6046f718 error 4 in libc-2.17.so[7fd86cf6e000+1c4000]
Jul  9 11:56:48 dejavu kernel: [813059.100894] cvm-unix[10872]: segfault at 0 ip 00007fde1c12b037 sp 00007ffdb360c128 error 4 in libc-2.17.so[7fde1bfec000+1c4000]
Jul  9 11:56:48 dejavu kernel: cvm-unix[10872]: segfault at 0 ip 00007fde1c12b037 sp 00007ffdb360c128 error 4 in libc-2.17.so[7fde1bfec000+1c4000]
Jul  9 12:13:40 dejavu kernel: [814071.287479] cvm-unix[10893]: segfault at 0 ip 00007fd95b990037 sp 00007ffd46252eb8 error 4 in libc-2.17.so[7fd95b851000+1c4000]
Jul  9 12:13:40 dejavu kernel: cvm-unix[10893]: segfault at 0 ip 00007fd95b990037 sp 00007ffd46252eb8 error 4 in libc-2.17.so[7fd95b851000+1c4000]
Jul  9 13:49:07 dejavu kernel: [819797.795402] cvm-unix[11439]: segfault at 0 ip 00007f0df2862037 sp 00007ffed5c3f878 error 4 in libc-2.17.so[7f0df2723000+1c4000]
Jul  9 13:49:07 dejavu kernel: cvm-unix[11439]: segfault at 0 ip 00007f0df2862037 sp 00007ffed5c3f878 error 4 in libc-2.17.so[7f0df2723000+1c4000]

That's alarming, though I don't know what to do with it...

I'll keep exploring logs...
« Last Edit: July 09, 2023, 10:27:48 PM by julianop »

Offline ReetP

  • *
  • 3,704
  • +5/-0
Re: where did the backup configuration page go?
« Reply #6 on: July 09, 2023, 10:41:05 PM »
So it was an upgrade from v9 - v10

How? Please explain as there are several methods.

Other issurs for reference.

Reboot/reconfigure
https://forums.koozali.org/index.php/topic,55004.0.html

SSL certs
https://forums.koozali.org/index.php/topic,55006.0.html

I'd suggest you really need to clear these up first as they may well be related.

You also had v9 backup issues a while ago too?

https://forums.koozali.org/index.php/topic,54032.0.html

Assuming this was the upgrade/restore method, had you tested the upgrade?

Is there anything in any of your logs that shows an error? I would expect something.

I suspect your problem goes back further than is currently apparent. Hence we need to rewind.

You should still be able to run a backup from a console/terminal as admin. Please check that.

But note, backing up an already corrupted system may not help you, no matter how desperate you are for it to work.

...
1. Read the Manual
2. Read the Wiki
3. Don't ask for support on Unsupported versions of software
4. I have a job, wife, and kids and do this in my spare time. If you want something fixed, please help.

Bugs are easier than you think: http://wiki.contribs.org/Bugzilla_Help

If you love SME and don't want to lose it, join in: http://wiki.contribs.org/Koozali_Foundation

Offline julianop

  • *
  • 61
  • +0/-0
Re: where did the backup configuration page go?
« Reply #7 on: July 09, 2023, 10:58:04 PM »
All that read and understood, ReetP, thanks.
Thanks for reminding me about the challenges back in '19; that was with an earlier version, and seemed to have been resolved. Occasionally I got errors, and had to start up a new GFS set on the backup NAS, but at least it was working.
With the fresh install of 10, I am mystified to find out that I don't seem to have set up backups at all, as the area on the NAS that I had created for it is completely empty. I have no understanding as to how that could have happened, other than to say that last year was very traumatic and difficult career-wise, and I may simply have been overwhelmed by events, and been interrupted before I completed the work.

Anyway, at this point, it seems, as you say, that the issues go back further than is an apparent, so I need a deep dive....

The only thing that matters for backup is email data, as everything else can be regenerated, so I'll do that now...

I'm tempted to do a simple rsync at this point, rather than attempt a tar; do you have any other guidance?

Thanks again !!

Offline TerryF

  • grumpy old man
  • *
  • 1,819
  • +6/-0
Re: where did the backup configuration page go?
« Reply #8 on: July 09, 2023, 11:25:01 PM »
No, it's no longer there; that's my point.
I realize that it should be there, but it's not. And no, I haven't been doing "something else" :-)

Sorry, didnt mean to be curt, just that it would appear it wasnt the install of the contrib that was causing the issue..
--
qui scribit bis legit

Offline julianop

  • *
  • 61
  • +0/-0
Re: where did the backup configuration page go?
« Reply #9 on: July 09, 2023, 11:54:15 PM »
No problem at all, TerryF, sorry if I seemed like I was triggered :-)

I appreciate all the help I get, and on behalf of others, I thank you and all others who help here for your patience and knowledge.

With millions of combinations of features and configurations, this stuff is difficult enough at the best of times, and with the passage of time it ain't getting any easier...

Here's an important update:
I found at least HALF of the missing backups problem; I don't know if it holds a clue to the missing Backups menu item...
It seems that when I did the version 10 install, I got interrupted. I had created a new share for the new version 10 backup archive on my NAS (just to avoid any version-related problems), but I had not installed nfs-utils to allow the system to connect to the nfs share, so even if I had configured backups, it couldn't have worked. If I remember from 9.1, the backup configuration dialog allowed me to select an nfs share as the target. Do I remember correctly?
Anyway, I have just installed nfs-utils, and I can manually mount the new nfs share on my NAS allocated to backups of the version 10 install and its accumulated data.
I can't see how the missing nfs support would cause the backup menu item to disappear, but I thought I'd mention it for the sake of completeness.

« Last Edit: July 10, 2023, 12:07:05 AM by julianop »

Offline julianop

  • *
  • 61
  • +0/-0
Re: where did the backup configuration page go? Solved.
« Reply #10 on: July 10, 2023, 02:22:43 AM »
William of Occam is still alive - probably good buddies with Mr Murphy.

So it seems that the reason that the backup config page wasn't available was that the "e-smith-backups" module simply wasn't installed. The menu item didn't disappear: it wasn't there in the first place, but because I was evidently interrupted during the install (starting a new job), I didn't notice.

On a Google search for "backups with koozali" this afternoon I found the link to:

 "https://wiki.koozali.org/index.php?title=SME_Server:10.0"

I found the term "Backups" in the changelog on that page, and it referred to updates in "e-smith-backups". Deducing that was a reference to an installable module, I attempt to
Code: [Select]
yum install it, and it installed. A refresh of Server-Manager brought the missing menu item back, and I was able to configure backups as before. And yes, I was of course able to select an nfs share as the target during the configuration.

Sometimes the simple hypothesis is the solution to the problem, and perhaps I should have thought to check, but shouldn't the backup module have been included in the standard install??


Offline Jean-Philippe Pialasse

  • *
  • 2,712
  • +11/-0
  • aka Unnilennium
    • http://smeserver.pialasse.com
Re: where did the backup configuration page go?
« Reply #11 on: July 10, 2023, 06:17:35 AM »
what gives

 grep e-smith-backup -r /var/log/yum/
 
 grep e-smith -r /root/anaconda*

Offline julianop

  • *
  • 61
  • +0/-0
Re: where did the backup configuration page go?
« Reply #12 on: July 10, 2023, 07:52:14 AM »
Code: [Select]
[root@dejavu files]# grep e-smith-backup -r /var/log/yum
/var/log/yum/yum.log.1653992659:---> Package e-smith-backup.noarch 0:2.6.0-27.el7.sme will be updated
/var/log/yum/yum.log.1653992659:---> Package e-smith-backup.noarch 0:2.6.0-28.el7.sme will be an update
/var/log/yum/yum.log.1653992659: e-smith-backup         noarch 2.6.0-28.el7.sme                smeupdates 122 k
/var/log/yum/yum.log.1653992659:  Updating   : e-smith-backup-2.6.0-28.el7.sme.noarch                   307/628
/var/log/yum/yum.log.1653992659:  Cleanup    : e-smith-backup-2.6.0-27.el7.sme.noarch                   355/628
/var/log/yum/yum.log-20230101:May 31 05:28:56 Updated: e-smith-backup-2.6.0-28.el7.sme.noarch
/var/log/yum/yum.log-20230101:Jun 05 01:21:44 Erased: e-smith-backup-2.6.0-28.el7.sme.noarch
/var/log/yum/yum.log:Jul 09 18:54:17 Installed: e-smith-backup-2.6.0-28.el7.sme.noarch
[root@dejavu files]#
[root@dejavu files]# grep e-smith -r /root/anaconda*
[root@dejavu files]#

and no, I have no idea why it was erased on Jun5th.... it certainly wasn't by my hand...

I did find this, though...
Code: [Select]
[root@dejavu files]# grep smeadmin -r /var/log/yum
/var/log/yum/yum.log:Jun 19 09:25:55 Installed: smeserver-smeadmin-1.6-10.el7.sme.noarch
[root@dejavu files]# grep "Jun 05" -r /var/log/yum
/var/log/yum/yum.log-20230101:Jun 05 01:21:44 Erased: smeserver-nfs-1.2.0-20.el7.sme.noarch
/var/log/yum/yum.log-20230101:Jun 05 01:21:44 Erased: e-smith-quota-2.6.0-5.el7.sme.noarch
/var/log/yum/yum.log-20230101:Jun 05 01:21:44 Erased: e-smith-flexbackup-2.6.0-3.el7.sme.noarch
/var/log/yum/yum.log-20230101:Jun 05 01:21:44 Erased: e-smith-backup-2.6.0-28.el7.sme.noarch
/var/log/yum/yum.log-20230101:Jun 05 01:21:44 Erased: 1:nfs-utils-1.3.0-0.68.el7.2.x86_64
/var/log/yum/yum.log-20230101:Jun 05 01:21:44 Erased: 1:quota-4.01-19.el7.x86_64
[root@dejavu files]#

Again, certainly nothing I did.
« Last Edit: July 10, 2023, 08:05:54 AM by julianop »

Offline ReetP

  • *
  • 3,704
  • +5/-0
Re: where did the backup configuration page go?
« Reply #13 on: July 10, 2023, 09:00:22 AM »
Quote
it certainly wasn't by my hand...

Again, certainly nothing I did.

IME these phrases are best left out of a debugging conversation as they usually prove incorrect....

Note you haven't described your restore to v10, and please explain this in more detail:

Quote
It seems that when I did the version 10 install, I got interrupted.

Before, during, after restore?

Roughly when did you upgrade?

I'd imagine this is at the root of your issues and has been compounded by your subsequent actions. Ignoring errors like this rarely turn out well as you are discovering.....

So what happened between these two?

Code: [Select]
/var/log/yum/yum.log-20230101:May 31 05:28:56 Updated: e-smith-backup-2.6.0-28.el7.sme.noarch
/var/log/yum/yum.log-20230101:Jun 05 01:21:44 Erased: e-smith-backup-2.6.0-28.el7.sme.noarch

Or what happened prior to Jun 05 at 01.21?

Something ran that erased the backup package.
...
1. Read the Manual
2. Read the Wiki
3. Don't ask for support on Unsupported versions of software
4. I have a job, wife, and kids and do this in my spare time. If you want something fixed, please help.

Bugs are easier than you think: http://wiki.contribs.org/Bugzilla_Help

If you love SME and don't want to lose it, join in: http://wiki.contribs.org/Koozali_Foundation

Offline julianop

  • *
  • 61
  • +0/-0
Re: where did the backup configuration page go?
« Reply #14 on: July 10, 2023, 11:41:24 AM »
Phrases: yeah, true... I'll meet you halfway by adding "knowingly, consciously or by explicit intent", OK? :-)

When did you upgrade?: Indications (file timestamps in places like /etc and /usr/bin on the new server, my last backup of the previous server, and some file/directory timestamps on the old server) are that I performed the upgrade over May 30th to May 31st 2022.
How...
I backed up my old server (9.2, "spencer") and performed a fresh install of Koozali 10 on a different computer ("dejavu") with new drives.
As the new machine is explicitly a mail server, I would have restored user files (Maildir) from /home/e-smith/files/users somehow. I'll address that part in a sec...

Interruption: The new server has been operating since the install, so the basic install  was obviously successful. The file restore opens up a question, though:... As I said in a previous post in this thread, the location that I created on my backup NAS for the new server is completely empty of backup files, so I obviously hadn't set up backups. When I tried to run a backup manually yesterday, I found that I couldn't mount the nfs drive, because nfs-utils wasn't installed. The lack of any backup data for dejavu, plus the absence of nfs-utils tells me that I couldn't have set up backups, although that would have been important to me. That's what I mean by "interrupted".
BUT somehow the Maildir contents for all the users was migrated to the new server, because we all have our old mail. There are three ways I know I could have done that:
1. via a tar restore;
2. by mapping the /home/e-smith/files/users directory as an nfs share into the new machine, and copying the files that way. The "exports" file on the old server is empty, and dated Jan 12 2010, so I cannot have moved files that way. /etc/samba/smb.conf is dated 31st May 2022, so i may have transferred some file that way, but the only shares in smb.conf are in ibays, which is not where Maildirs are located.
3. By setting up imap accounts for each of my users on both the old and new servers on my email client, and using that to copy all the mail files from one account to the old server to the corresponding account on the new server. Because I only needed to transfer mail, that's probably the method I used.

"What happened between these two?" Nothing, to my knowledge: as I said, there's nothing that I would have done myself. And there is absolutely no reason for me to have removed the e-smith-backup module. I didn't even know about that module at that time, and while, because of work stress at the time (new job) I may have forgotten to set up backups, the chances of my having inadvertently removed the very module responsible for doing it are beyond improbable. And to remove nfs as well? Something that I knew perfectly well that I would need for backups? That is simply outside the realms of plausibility. No, I insist that I did not remove those modules myself.

"Prior to June 5th" Well that's this year... I wouldn't have been doing anything with the server at that time: I was preparing for a business visit to my head office around that time, and I left town June 4th for a week.
Here's an extract from the update log around June 5th...
Code: [Select]
/var/log/yum/yum-update.log-20230524:May 24 00:19:43 dejavu yum_update_dbs: Existing lock /var/run/yum.pid: another copy is running as pid 3488.
/var/log/yum/yum-update.log-20230524:May 24 00:19:43 dejavu yum_update_dbs: Another app is currently holding the yum lock; waiting for it to exit...
/var/log/yum/yum-update.log-20230524:May 24 00:19:43 dejavu yum_update_dbs: The other application is: yum
/var/log/yum/yum-update.log-20230524:May 24 00:19:43 dejavu yum_update_dbs: Memory : 112 M RSS (1.4 GB VSZ)
/var/log/yum/yum-update.log-20230524:May 24 00:19:43 dejavu yum_update_dbs: Started: Wed May 24 00:18:50 2023 - 00:53 ago
/var/log/yum/yum-update.log-20230524:May 24 00:19:43 dejavu yum_update_dbs: State  : Uninterruptible, pid: 3488
/var/log/yum/yum-update.log-20230606:Jun  5 03:47:20 dejavu yum_update_dbs: BDB2053 Freeing read locks for locker 0x251: 32495/140502041245504
/var/log/yum/yum-update.log-20230606:Jun  5 03:47:20 dejavu yum_update_dbs: BDB2053 Freeing read locks for locker 0x253: 32495/140502041245504
/var/log/yum/yum-update.log-20230606:Jun  5 03:47:20 dejavu yum_update_dbs: BDB2053 Freeing read locks for locker 0x262: 32495/140502041245504
/var/log/yum/yum-update.log-20230606:Jun  5 03:47:20 dejavu yum_update_dbs: BDB2053 Freeing read locks for locker 0x264: 32495/140502041245504
/var/log/yum/yum-update.log-20230606:Jun  5 03:47:20 dejavu yum_update_dbs: BDB2053 Freeing read locks for locker 0x265: 32495/140502041245504
/var/log/yum/yum-update.log-20230606:Jun  5 03:47:20 dejavu yum_update_dbs: BDB2053 Freeing read locks for locker 0x266: 32495/140502041245504
/var/log/yum/yum-update.log-20230606:Jun  5 03:47:20 dejavu yum_update_dbs: BDB2053 Freeing read locks for locker 0x267: 32495/140502041245504
/var/log/yum/yum-update.log-20230606:Jun  5 03:47:20 dejavu yum_update_dbs: BDB2053 Freeing read locks for locker 0x268: 32495/140502041245504
/var/log/yum/yum-update.log-20230606:Jun  5 03:47:20 dejavu yum_update_dbs: BDB2053 Freeing read locks for locker 0x269: 32495/140502041245504
/var/log/yum/yum-update.log-20230606:Jun  5 03:47:20 dejavu yum_update_dbs: BDB2053 Freeing read locks for locker 0x26a: 32495/140502041245504
/var/log/yum/yum-update.log-20230606:Jun  5 03:47:20 dejavu yum_update_dbs: BDB2053 Freeing read locks for locker 0x26b: 32495/140502041245504
/var/log/yum/yum-update.log-20230606:Jun  5 03:47:20 dejavu yum_update_dbs: BDB2053 Freeing read locks for locker 0x26c: 32495/140502041245504
/var/log/yum/yum-update.log-20230606:Jun  5 03:47:20 dejavu yum_update_dbs: BDB2053 Freeing read locks for locker 0x26d: 32495/140502041245504
/var/log/yum/yum-update.log-20230606:Jun  5 03:47:20 dejavu yum_update_dbs: BDB2053 Freeing read locks for locker 0x26e: 32495/140502041245504
/var/log/yum/yum-update.log-20230606:Jun  5 03:47:20 dejavu yum_update_dbs: BDB2053 Freeing read locks for locker 0x26f: 32495/140502041245504
/var/log/yum/yum.log.1687183069:Warning: RPMDB altered outside of yum.
/var/log/yum/yum.log.1687183069:Reload yum db for server-manager
/var/log/yum/yum-update.log-20230619:Jun 19 08:59:21 dejavu yum_update_dbs: Existing lock /var/run/yum.pid: another copy is running as pid 27150.
/var/log/yum/yum-update.log-20230619:Jun 19 08:59:21 dejavu yum_update_dbs: Another app is currently holding the yum lock; waiting for it to exit...

As if that wasn't bad enough, I see this excerpt from yesterday in /var/log/messages...
Code: [Select]
Jul 10 01:00:01 dejavu su: (to qmailr) root on none
Jul 10 01:04:17 dejavu kernel: [860305.626185] cvm-unix[32656]: segfault at 0 ip 00007f7c015b3037 sp 00007ffeb5b4a578 error 4 in libc-2.17.so[7f7c01474000+1c4000]
Jul 10 01:04:17 dejavu kernel: cvm-unix[32656]: segfault at 0 ip 00007f7c015b3037 sp 00007ffeb5b4a578 error 4 in libc-2.17.so[7f7c01474000+1c4000]
Jul 10 01:05:01 dejavu su: (to qmailr) root on none
Jul 10 01:10:01 dejavu su: (to qmailr) root on none
Jul 10 01:15:01 dejavu su: (to qmailr) root on none

I'm wondering if, at this point, I shouldn't just do a fresh install..

Offline ReetP

  • *
  • 3,704
  • +5/-0
Re: where did the backup configuration page go?
« Reply #15 on: July 10, 2023, 02:51:43 PM »
Nope none of it makes any real sense. I feel we have an XY Problem here with only half the info.

Wood from trees.

Quote
Jul 10 01:04:17 dejavu kernel: [860305.626185] cvm-unix[32656]: segfault at 0 ip 00007f7c015b3037 sp 00007ffeb5b4a578 error 4 in libc-2.17.so[7f7c01474000+1c4000]
Jul 10 01:04:17 dejavu kernel: cvm-unix[32656]: segfault at 0 ip 00007f7c015b3037 sp 00007ffeb5b4a578 error 4 in libc-2.17.so[7f7c01474000+1c4000]

This is not relevant. There is a bug which causes cvm-unix to restart.

cvm-unix fails, and correctly restarts itself. As cvm-unix is very old an unmaintained we are moving to use the auth_imap plugin which seems to work OK.

You can test it with this updated package.

Code: [Select]
yum --enablerepo=smeupdates-testing install smeserver-qpsmtpd
Upgrade

Quote
upgrade over May 30th to May 31st 2022

Definitely 2022 right? Just to be sure?

Backups.

Very odd here and surprised you don't remember how you restored. That complicates matters. A standard backup to USB/Restore works without issues.

Anything else - YMMV. Still not sure that copying over some mail files will create the sequence of events you state have occurred (reconfigure/reboot - Certificates - backups)


Logs - they don't appear to be complete. Wondering what made this lot remove themselves? What were the lines preceding this? Did you do an upgrade? Accidental removal?

Code: [Select]
/var/log/yum/yum.log-20230101:Jun 05 01:21:44 Erased: smeserver-nfs-1.2.0-20.el7.sme.noarch
/var/log/yum/yum.log-20230101:Jun 05 01:21:44 Erased: e-smith-quota-2.6.0-5.el7.sme.noarch
/var/log/yum/yum.log-20230101:Jun 05 01:21:44 Erased: e-smith-flexbackup-2.6.0-3.el7.sme.noarch
/var/log/yum/yum.log-20230101:Jun 05 01:21:44 Erased: e-smith-backup-2.6.0-28.el7.sme.noarch
/var/log/yum/yum.log-20230101:Jun 05 01:21:44 Erased: 1:nfs-utils-1.3.0-0.68.el7.2.x86_64
/var/log/yum/yum.log-20230101:Jun 05 01:21:44 Erased: 1:quota-4.01-19.el7.x86_64

Particularly like to see all of this one:

Code: [Select]
/var/log/yum/yum.log-20230101
And all of the /var/log/messages log for the same June 04/05

Thing is you probably did an upgrade around May 31 to get this:

/var/log/yum/yum.log-20230101:May 31 05:28:56 Updated: e-smith-backup-2.6.0-28.el7.sme.noarch

But it was removed on Jun 05

/var/log/yum/yum.log-20230101:Jun 05 01:21:44 Erased: e-smith-backup-2.6.0-28.el7.sme.noarch

So what happened Jun 04/05? You say you were away, but Koozali SME does not auto update or autoremove AFAIA.

I have been trying to see what could have been removed to cause the packages listed above to be removed. At a guess at least nfs-utils, but removing quota is a tricky one to understand.

I also note that there had been an update to smeserver-nfs-1.2.0-20.el7.sme.noarch last November - I doubt that your version worked very well.

Code: [Select]
%changelog
* Fri Nov 18 2022 Jean-Philippe Pialasse <tests@pialasse.com> %{version}-21.sme
- fix service not starting and not accessible [SME: 12230]
- using systemd and removing sysvinit links

So you probably had no backups since you installed it because nfs wouldn't start. When was that? Did you ever check the mappings and settings when you installed it?

How about looking at these:

Code: [Select]
yum history
history |grep rpm

Sorry but there are just so many unknowns here to be able to tell anything definitively.
...
1. Read the Manual
2. Read the Wiki
3. Don't ask for support on Unsupported versions of software
4. I have a job, wife, and kids and do this in my spare time. If you want something fixed, please help.

Bugs are easier than you think: http://wiki.contribs.org/Bugzilla_Help

If you love SME and don't want to lose it, join in: http://wiki.contribs.org/Koozali_Foundation

Offline julianop

  • *
  • 61
  • +0/-0
Re: where did the backup configuration page go?
« Reply #16 on: July 10, 2023, 03:32:15 PM »
No sense: I agree that it doesn't appear to. And yes, I'm sure it is a half the info issue. Plase be certain that I am not intentionally leaving anything out. I respect your time.

Segfault: thanks, then I'll ignore it. Per note abve, I didn't want to leave anything else!

Upgrade over May 30th to May 31st 2022: In the immortal words of Father Cornelius (Fifth Element): "Absolutely sure!". There's plenty of evidence for that.

Backups/restore: Yes, I see how that would that might surprise you :-) For one thing, I'm 67, so my memory isn't quite as good as it used to be. Second, This isn't my day job, so while I've been running Linux systems at home for what must now be two decades, it is not my job, but a hobby. I'm an embedded systems programmer by trade, since '82.

Backup to USB: Er, no... My backup would have been a workstation backup to a NAS; I have 36G of historical email data.....  I did a completely fresh install to a different machine. The old one is still sitting on my workbench in my basement, with nothing to do but talk to itself...

YMMV: agreed. I don't remember exactly when I installed the certificate manager, perhaps it was to assist in this year's cert update effort, which I was a few days late in performing; I'll look through the logs for evidence. I installed Smemanager only very recently.

My workday started 25 minutes ago, so I'll end here for now, and answer the reset of your questions, including the requests for data, throughout the day. My postings that are in response to your requests are not edited in any way, they are verbatim copy/pastes.

Meanwhile: It is in my interests to resolve this, so I will do everything I can within my power to help you to help me. I really appreciate your time, and will not abuse it.

Back soon...

Thank you :-)

Offline ReetP

  • *
  • 3,704
  • +5/-0
Re: where did the backup configuration page go?
« Reply #17 on: July 12, 2023, 04:58:00 PM »
Quote
This isn't my day job

See 4. in my sig.

This isn't our jobs either. Everyone who codes here is a volunteer and everyone does it in their spare time.
...
1. Read the Manual
2. Read the Wiki
3. Don't ask for support on Unsupported versions of software
4. I have a job, wife, and kids and do this in my spare time. If you want something fixed, please help.

Bugs are easier than you think: http://wiki.contribs.org/Bugzilla_Help

If you love SME and don't want to lose it, join in: http://wiki.contribs.org/Koozali_Foundation

Offline julianop

  • *
  • 61
  • +0/-0
Re: where did the backup configuration page go?
« Reply #18 on: July 12, 2023, 10:29:37 PM »
Exactly. You're heroes :-)

Offline ReetP

  • *
  • 3,704
  • +5/-0
Re: where did the backup configuration page go?
« Reply #19 on: July 13, 2023, 12:36:26 AM »
We can't help fix stuff without data.......
...
1. Read the Manual
2. Read the Wiki
3. Don't ask for support on Unsupported versions of software
4. I have a job, wife, and kids and do this in my spare time. If you want something fixed, please help.

Bugs are easier than you think: http://wiki.contribs.org/Bugzilla_Help

If you love SME and don't want to lose it, join in: http://wiki.contribs.org/Koozali_Foundation