Topic: affa 3 issues

[jameswilson]

I have a backup machine running affa and all is working well on my servers but a friends system I'm having issues with.
It started life as a 7 machine upgraded to 8 then 9 with affa

However I cant seem to back it up anymore

[root@bu16server affa]# affa --configcheck
Checking job section GlobalAffaConfig (/etc/affa/GlobalAffaConfig.conf)
Checking job section atautos5 (/etc/affa/atauto5.conf)
Checking job section localhost (/etc/affa/localhost.conf)
Checking job section secware5 (/etc/affa/secware5.conf)
Checking job section siaholdings5 (/etc/affa/siaholdings5.conf)
Checking job section tsi5 (/etc/affa/tsi5.conf)
Configuration  is ok.
[root@bu16server affa]# affa --send-key atautos5

Job atautos5: root@atautomotive.co.uk's password:
Public key sent to atautomotive.co.uk
[root@bu16server affa]# affa --run atautos5

[root@bu16server affa]#

all seems ok but

[root@bu16server ~]# affa -sta
Affa version 3.2.2.3-2 on bu16server
+--------------+-----+-------+--------+-------+-------+-------+----------------+
| Job          | ENA |  Last |   Time |  Next |  Size | ddYld | N of S,D,W,M,Y |
+--------------+-----+-------+--------+-------+-------+-------+----------------+
| atautos5     | yes | waiting (pid 3191)                     |  0, 0, 0, 0, 0 |
| secware5     | yes | running rsync (pid 2879)               |  0, 0, 0, 0, 0 |
| siaholdings5 | yes | running rsync (pid 2468)               |  0, 0, 0, 0, 0 |
| tsi5         | yes | running rsync (pid 2764)               |  0, 0, 0, 0, 0 |
+--------------+-----+-------+--------+-------+-------+-------+----------------+
1 disabled jobs not listed. Use --all to display.

Then it fails

Any suggestions as to what I'm doing wrong?

Ta
James

[Stefano]

Well, you forgot to post here the error message..

[jameswilson]

lol
I don't get one at the prompt?

[Stefano]

Well, check logs

[jameswilson]

Thanks
I have before but the error I get is about the public key

Sat Sep 10 19:37:07[3353]: Affa 3.2.2.3-2: Running /sbin/affa --run atautos5 --RetryAttempts=0 --RetryAfter=600
Sat Sep 10 19:37:07[3353]: ##################
Sat Sep 10 19:37:07[3353]: # Affa 3.2.2.3-2 #
Sat Sep 10 19:37:07[3353]: ##################
Sat Sep 10 19:37:07[3353]: Starting job atautos5 scheduled (atauto???.co.uk)
Sat Sep 10 19:37:07[3353]: Description: Backup of ATautos by 192.168.16.241 virtual machine
Sat Sep 10 19:37:07[3353]: Waiting 600 seconds. Continuing at 19:47:07
Sat Sep 10 19:47:07[3353]: Checking SSH connection to root@atauto???.co.uk
Sat Sep 10 19:48:10[3353]: Error (1260): SSH connection to atauto???.co.uk failed. Did you send the public key?
Sat Sep 10 19:48:10[3353]: Total execution time: 11m03s
Sat Sep 10 19:48:10[3353]: Failure message sent to root
Sat Sep 10 19:48:10[3353]: Exiting.
Sat Sep 10 19:48:10[3353]: .

but as you can see I have sent it previously and since built a new vm in case it was that?
Thanks for looking though. I have edited the full url

James

[Stefano]

Try to connect via SSH to the target machine and see what happen
Try to check logs on target

[jameswilson]

Many thanks

I can ssh in fine
WHat logs am I looking at please?
There is nothing in messages

[Stefano]

No affa logs?

[jameswilson]

On the target, ill look I only looked in messages.

I have noticed that when trying to start the job manually it doesn't ask me the yes/no question about certificates etc. But send key and check connections shows all is ok.

[jameswilson]

No affa logs on the target machine

[guest22]

Any chance you have an expired ssh key on any of the machines? Try delete all keys and start again.

[jameswilson]

maybe where are the keys located?
I first installed my mates system on sme7 in 2006

[guest22]

Anything in /root/.ssh ?

[jameswilson]

ive deleted authorized_keys2 and

affa -send-key etc

now it asks (which it hasn't before)

affa -run atautos5
The authenticity of host 'atautos5 ([213.254.185.201]:1977)' can't be established.
RSA key fingerprint is cd:a5:20:96:d4:90:f3:ab:88:4c:69:41:92:1c:65:6d.
Are you sure you want to continue connecting (yes/no)? yes

on the backup machine

[root@bu16server ~]# affa -st
Affa version 3.2.2.3-2 on bu16server
+--------------+-----+-------+--------+-------+-------+-------+----------------+
| Job          | ENA |  Last |   Time |  Next |  Size | ddYld | N of S,D,W,M,Y |
+--------------+-----+-------+--------+-------+-------+-------+----------------+
| atautos5     | yes | waiting (pid 13187)                    |  0, 0, 0, 0, 0 |
| barrybufton5 | yes | running rsync (pid 12774)              |  0, 0, 0, 0, 0 |
| secware5     | yes | 05:18 |  3m31s | 05:15 |   34G |     - |  1, 1, 0, 0, 0 |
| siaholdings5 | yes | 02:18 |  3m33s | 02:15 |  460G |     - |  1, 1, 0, 0, 0 |
| tsi5         | yes | 05:55 | 11h05m | 04:15 |   42G |     - |  1, 0, 0, 0, 0 |
| wss5         | yes | running rsync (pid 12962)              |  0, 0, 0, 0, 0 |
+--------------+-----+-------+--------+-------+-------+-------+----------------+
1 disabled jobs not listed. Use --all to display.

I don't know what its waiting for?

looking at the log I have found this

Sun Sep 11 14:07:03[13108]: Affa 3.2.2.3-2: Running /sbin/affa -run atautos5
Sun Sep 11 14:07:03[13108]: ##################
Sun Sep 11 14:07:03[13108]: # Affa 3.2.2.3-2 #
Sun Sep 11 14:07:03[13108]: ##################
Sun Sep 11 14:07:03[13108]: Starting job atautos5 scheduled (atautomotive.co.uk)
Sun Sep 11 14:07:03[13108]: Description: Backup of ATautos by 192.168.16.241 virtual machine
Sun Sep 11 14:07:03[13108]: Checking SSH connection to root@atautomotive.co.uk
Sun Sep 11 14:07:04[13108]: Installing watchdog on atautomotive.co.uk with dt=87000 seconds
Sun Sep 11 14:08:10[13108]: Error (3508): Couldn't copy /tmp/13108.affa-watchdog-atautos5-192.168.16.241 to remote host.
Sun Sep 11 14:08:10[13108]: Starting re-run 1 of 4 in 600 seconds.
Sun Sep 11 14:08:10[13108]: Total execution time:  1m07s
Sun Sep 11 14:08:10[13108]: Failure message sent to root
Sun Sep 11 14:08:10[13108]: Exiting.

I'm assuming

Sun Sep 11 14:08:10[13108]: Error (3508): Couldn't copy /tmp/13108.affa-watchdog-atautos5-192.168.16.241 to remote host.

I'm assuming this is the culprit now the certificate thing is sorted?

[guest22]

who owns /tmp ?