Confirm successful reboot of linux servers

[lubu]

Hi All,

I would like to know how to confirm successful reboot of linux servers.
I have to remotely monitor the reboots of many servers. When I get a message that a unix/linux server is rebooting I issue a continuous ping and if the server goes down and comes back up I assume that the reboot was successful.

In the case of windows servers, I check the event viewer for event ID 6005, 6009 and 6006 to confirm the success reboot.

Is there a similar procedure I can follow for linux/unix servers?

Please advise...

Thanks

Lubu

[DingoBoy]

Hi All,

I would like to know how to confirm successful reboot of linux servers.
I have to remotely monitor the reboots of many servers. When I get a message that a unix/linux server is rebooting I issue a continuous ping and if the server goes down and comes back up I assume that the reboot was successful.

In the case of windows servers, I check the event viewer for event ID 6005, 6009 and 6006 to confirm the success reboot.

Is there a similar procedure I can follow for linux/unix servers?

Please advise...

Thanks

Lubu

Hmm. Do you know of the program wall? I am not entirely sure it could help you, but i believe it would be possible to have the server in question tell you when it reboots. (I saw nothing in the manpage for wall that would imply it only works on the one system, but nothing contrary either - can anyone confirm if this would be possible?)

Another solution could theoretically be echo'ing lines of text from the server in question to you PC's bash, however i know nothing of how this could be practically implemented.

Hope my ideas were of any use.

Comradely, DingoBoy.

[Freston]

Welcome to the forums!

I don't have experience with Linux servers in a live professional environment, but some thoughts and things that work for me are...


NFS - one system I use atm for multiple clients that are being tested and that works pretty well for me, the clients write status rapports to a logfile that sits on an exported nfs directory. As always with nfs, securety is a point of attention.
SSH - Systems can report their status through ssh. You could make a small script that requests status rapports from certain or all servers. My home server gives a status rapport this way with only a click on an icon, very convenient. But also, the server can give rapports to a central place without request. You need to use passwordless authentification for this, so security is a point of attention.
nc - have the servers report on a central nc channel that writes to a logfile. I think this is the safest option, because you can easily set it up to receive everything as flat text and read it as such.
There are other possibilities too, but these seem pretty straightforward.

Just add whatever the server needs to rapport during boot in the boot scripts.

[lubu]

Thanks DingoBoy and Freston for the suggestions

The ssh option is something I should be able to implement. Any other ideas are always welcome