Topic: what is this sme box used for?

[winecooler]

So I have a couple of other active threads going.... put simply the old tech who installed this back up sme server is gone and I want to know what it was backing up. I believe it was windows server data, unsure if it was a full server backup or workstations or a share folder or ....?

What I need is to hire a new tech with sme or linux expertise but the win server has crashed and I need to see if there is a full backup on this sme server, it's been 2 days and the staff are all standing around with pitchforks.

I have limited linux expertise but I can "cd" and "ls" my way around. server-manager is not working, but to fix that I have to fix the certificate issue, and that's not working... so I'm hoping someone can show me:

1. where to look for volumes on that machine that might contain backups, if any
2. where to look for what kind of backup software package might have been used, so if need be I can use it to restore
3. is it even possible for sme server to back up a baremetal windows server? Or is it likely it was used just to back up user data?

I will continue to pursue the technical fixes to the sme server issues so that at least I can run server-manager and have a look, but until then if anyone can offer any quicker answers to the above, it would be appreciated

[guest22]

pm me and I will help out.

Update: bummer, somebody forbids the use of PM on these forums.

[guest22]

on the console:

rpm -qa |grep affa

if it returns a positive feedback like an installed affa RPM:

- Backups are in /var/affa

see: http://wiki.contribs.org/Affa

[winecooler]

Thank you RequestedDeletion... there is no Affa on the machine. I think there's 2 hard drives on the machine, maybe there's some data on the 2nd one?

[CharlieBrady]

I think there's 2 hard drives on the machine, maybe there's some data on the 2nd one?

The two drives are likely being used in a RAID1 mirror pair for the SME server software and files.

[janet]

winecooler

So you have root access now to the command prompt after resetting the root & admin passwords, correct ?
At present you are NOT able to access the web GUI server manager using https://yourdomain/server-manager, correct ?
What error or message do you receive ?

Can you access the admin console menu from the command prompt ?:" ie type
console

If so, can you then access the text based server manager from the console admin menu ?
What happens or what error message do you get ?

To discover more about your server & how it is configured & potentially what it is doing,
here are a few commands to show us the output of:

df -h

cat /proc/mdstat

cat /proc/version

fdisk -l

/sbin/e-smith/audittools/newrpms

/sbin/e-smith/audittools/templates

/sbin/e-smith/audittools/repositories

This should tell us a lot about your server & what it is doing

[winecooler]

So you have root access now to the command prompt after resetting the root & admin passwords, correct ?
At present you are NOT able to access the web GUI server manager using https://yourdomain/server-manager, correct ?
What error or message do you receive ?

Yes I have root. Correct, I cannot access web gui, if I use http://... it gives me (111) connection refused, and if I use https://... I get "couldn't establish a connection to the server". I can ping the machine from that Windows workstation, so I know it's there.

Can you access the admin console menu from the command prompt ?:" ie type
console

If so, can you then access the text based server manager from the console admin menu ?
What happens or what error message do you get ?

Whaaaat, there's a text based server manager?!  Coolbeans.... server status is up, internet access = good, raid in a clean state, but when I try to access the text based server manager from there, it says "connection refused". This probably has something to do with its expired certificate issue?

To discover more about your server & how it is configured & potentially what it is doing,
here are a few commands to show us the output of:

These are handwritten notes carried to another machine, so bear with me if i leave out details, i can run them again if need be:

df -h
/dev/mapper/main-root 82% use% on /
/dev/md1 44% on /boot
none 0% on /dev/shm

cat /proc/mdstat
active raid1

cat /proc/version
Linux 2.6.9-89.31.1 ELsmp

fdisk -l
/dev/hda 320gb linux raid autodetect
/dev/md1 106mb
/dev/md2 319gb
/dev/dm-0 319gb
/dev/dm-1 536mb

/sbin/e-smith/audittools/newrpms
cannot find a valid baseurl for repo: base
plugins: installonlyn, smeserver, fastestsomething (can't read my writing lol), protect packages, not using ftp, http(s), or file for repos, skipping 4 is not a valid release

/sbin/e-smith/audittools/templates
..templates-custom/home/e-smith/ssl.crt: manually_added, override
.../etc/crontab/weeklybackup.cronjob manually added, addition
dailybackup.cronjob manually added, addition
disk space watchdog.cronjob manually added, addition
syncwindowsshare.cronjob manually added, addition
dailypurge.cronjob manually added, addition
monthlybackup.cronjob manually added, addition
/smb.conf/11serverstring.cronjob manually added, override

/sbin/e-smith/audittools/repositories
disabled: addons, centosplus, extras, smecontribs,smedev,smetest,smeupdates-testing
enabled: base, smeaddons, smeextras,smeos,smeupdates

The thing that sticks out to me is: syncwindowsshare.cronjob ... that might suggest its backing up windows shares?

[janet]

winecooler

cat /proc/version
Linux 2.6.9-89.31.1 ELsmp

That appears to be a sme7.x server

We do need the complete output details of those commands to see the whole picture.
Do these first
df -h
cat /proc/mdstat

Does midnight commander run ? type
mc

The thing that sticks out to me is: syncwindowsshare.cronjob ... that might suggest its backing up windows shares?

You can navigate the menus/folders using mc & read the file contents of the cron job(s) to see what they are doing
That will give some more clues

or at command prompt do

cat /etc/cron.daily/cronjobname
(put in the actual correct details)

I think you need to stabilize your machine before you investigate what it is doing backup wise.
See my queries in your certificate post

[winecooler]

okay here are some more pics. The cron jobs look like they are old, i'm guessing they came with the install.

[janet]

winecooler

/sbin/e-smith/audittools/newrpms
cannot find a valid baseurl for repo: base
plugins: installonlyn, smeserver, fastestsomething (can't read my writing lol), protect packages, not using ftp, http(s), or file for repos, skipping 4 is not a valid release

You are getting errors here & the command cannot complete, because sme7.6 went EOL (end of life) some time ago.
The repos had to be changed on the server.
There were updates to fix this IIRC
You need to reconfigure the repos correctly.

We then start getting into a lot of troubleshooting issues with a dysfunctional sme 7.6 server, and we do not know how long that will take & how much will need to be fixed.
We are not even sure if your sme7.x is really up to date ie does it have the last set of releases for sme7.6.

It may be better (less effort on our part) for you to upgrade to sme8 & sort out any problems you have then.
My guess is that many issues may resolve themselves.

Not sure if this is the wisest approach as it would be good to know what is functional in you server first, ie is the RAID really RAID1, are both drives alive or is the array degraded, ie only one drive functioning.
Do you have backups of the server if something went wrong during the sme7.x to sme8 upgrade.
Many questions ???

please show output of

db yum_repositories show base
db yum_repositories show smeupdates

What does this show
yum update

cancel the update by answering n to play safe for now

If that errors out too much, then do (& show us the output) we just need the list of rpms to be updated

yum update --disablerepo=base --disablerepo=smeupdates

also cancel with n to be safe for now

[janet]

winecooler

OK those pics show us you have 2 x320Gb drives hda & hdc in RAID1 array
The array is in sync & healthy (the UU part)

This all looks good.
 
show us a pic of
df -h

to check space usage, lack of space is known to stop http

[janet]

winecooler
 

show us a pic of
df -h

Oops sorry, I see that now.
53Gb available space
& 53M available om boot

so all that is OK too.

[winecooler]

I really appreciate your help, this board is lucky to have someone as responsive as you.
Although I understand what you are saying about fixing the problems/upgrading etc, all I really want to know is, what is the purpose of this server? If the server is doing nothing useful, then there's no point upgrading it, I'll just shut it down forever. Even if it is actually running some kind of backup, I'll just replace it with something I know asap.

I will get photos of the yum things you ask for if it will help

thanks again janet!

[janet]

winecooler

....what is the purpose of this server?
.....Even if it is actually running some kind of backup, I'll just replace it with something I know asap.

I cannot answer that at this stage, more info needed.
It would be very useful to get access to server manager to see if any backups are configured there.
There could be other systems/contribs/whatever running from command line cron jobs that do not even appear in server manager.
We need to get a proper listing of the rpms installed, so need to resolve the old repos issue, catch 22.
Read my earlier post queries & respond.

[winecooler]

yummy stuff