EV-SMTP

[WiTSend]

I have migrated to EV-SMTP journaling for my environment.  I have two questions:

1.  Is there an "in the box" solution to clearn up the SMTPHold empty folders?  I can see thousands of empty folders building up on each EV-SMTP server.  I suppose I can script deletion after so many days, but I'd rather have a built-in process.

2.  Is anyone out there using EV-SMTP and have a method of monitoring?  I have mulitple servers and would like some kind of dashboard/webapp that I can use to monitor the EV SMTP activity.

TIA,

[JimmyNeutron]

Hi,

Please leave these folders intact. EV will reuse them as it needs to. This is by design. I believe this was answered a while back by support and there is a thread about it somewhere. I'll try to dig it up and link you to it.

[JimmyNeutron]

Here is the link to the older thread regarding these empty folders.

https://www.veritas.com/community/forums/ev11-smtp-archiving-holding-folder-empty-folders#comment-11...

[dcVAST]

You can remove them but do not need to as EV will reuse them. For monitoring, you can of course monitor the size of the SMTP holding folder. You can also look at event IDs like 41547, which are usually given during backup mode or if there is no open partition etc. You can also monitor the mail queues, as there will be a build up if the messages are not being sent to the EV server.

[GertjanA]

For monitoring, there should be several in the SCOM pack.

also see http://www.veritas.com/docs/000024822 which has some Performance counters listed you can look at. Page 86 and further.

Also, from memory, a file will be created on the EV SMTP server (I believe in EV\reports\smtp, but not sure) which has some info.

Obviously, you need to closely monitor queues in Exchange, to prevent memory filling up if EV smtp is down.

There also is a EV powershell to check the service itself (in same document as listed above)

GJ

[WiTSend]

Thanks all for you comments.  I guess I'm left with developing my own dashboard.  With multiple EV servers trying to go to each report folder and server for perfmon is less than efficient.  We'll create something.