Jump to content

A suggestion to prevent problems...


Recommended Posts

Today i was checking a client's exchange server, where there was a mail store of just over 2.6 gig of a 10 gig drive. Retrospect's exchange agent threshold was set to 75% by default, so what happens?

 

 

 

0 bytes free on the mail servers store drive.

 

 

 

No matter where you are from, thats NOT a good thing =)

 

 

 

PLEASE change the criteria so that its default criteria doesn't put you in this situation, default the flushing to a day of the week, or month, or something. Having defaults which guarantee trouble later isn't very good.

 

 

 

Thanks

 

 

 

 

 

 

 

 

 

 

 

 

Link to comment
Share on other sites

It doesn't necessarily follow that the default behavior is bad. Dantz has no idea where the Exchange data will be stored, for instance; many administrators designate a separate partition to hold the stores. Also, a default flush based on an arbitrary interval might unacceptably tie up server processing cycles when people wouldn't want it to happen.

 

 

 

Essentially, we chose the most "vanilla" of defaults, because we simply can't predict how any given Exchange environment is going to be set up.

 

 

 

Having said that, though, we're examining ways that we can either make the Agent more intuitive, or provide better documentation for it up front.

Link to comment
Share on other sites

I don't think anyone is going to make a dedicated partition just for the retrospect log files. Filling any harddrive that is not dedicated to the logfiles is going to cause problem for other applications no matter what. A performance hit is definitely preferable to downtime, and there are always times when the load on a server is otherwise light (ie, late sunday night, early monday morning).

 

 

 

a safe vanilla configuration is what is needed, not one that will cause grief to your customers in most situations. If you don't want to do that, please put in a big warning so that during installation the end user knows and understands the issue before it has an impact on their servers.

 

 

 

For me the first time i knew about the issue was when it had already caused downtime, definitely not a good thing for my clients.

 

 

 

I hope there are no other issues that are going to raise their ugly heads ;)

 

 

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...