Yesterday everything was great. Today the search is taking more then 10 minutes. I am very new to exchange, and I am not even sure where to start looking. I can't reboot the server.
The problem is in both web and outlook
OutlookExchangeWindows Server 2012
Last Comment
Lofty Worm
8/22/2022 - Mon
JAGDIS S
Hello ,
i am ready to help you , first of all login into exchange server open run and type the command extra and follow the wizard and hare me the results .based on this report we can start troubleshooting the actual issue.
Regards
Scott C
Could be a performance issue.
First off I need some specific information.
What version of Exchange? This includes Service Packs and either CU or RU level
You can get this by running the command...."Get-Command exsetup | foreach {$_.FileVersionInfo}".
It's always best to be on the latest SP and at least n-1 CU / RU level.
Look at the report generated and see if you can see any disk latency or anything else in red that could affect disk performance.
Also make sure your servers are all patched up as well.
Is this happening for other users?
How many messages are in the mailbox/folder you are searching in?
Scott C
@JAGDIS S....how do you plan on reading the Extra traces? Those can only be read by Microsoft Engineers. I know this as I was a Tier 3 Exchange Engineer for 15 months.
In addition, you didn't supply the proper tags to select to troubleshoot this type of performance issue. And if all tags are selected, the trace will contain so much information as to be unusable.
TY,
The version is
ProductVersion FileVersion FileName
-------------- ----------- --------
15.00.1130.007 15.00.1130.007 C:\Program Files\Microsoft\Exchange Server\V15\bin\ExSetup.exe
Lofty Worm
ASKER
I am not permitted to load additional files on the server itself. Is there another way to gather this data?
Lofty Worm
ASKER
There is no command "extra" Am I missing something?
I may have a cluster node failure. Please stand by, more to come........
Scott C
Ok. You're on CU10. Not the latest but you should be ok. BTW...You're about a year out of date on this server.
"extra" does not come with Exchange 2013. It needs to be installed separately. And as I said, it should only be run under the direction of Microsoft Exchange engineers.
ExPerfWiz is a self-contained script supported by MS. No, not really. As you are new to Exchange, getting the correct counters would be next to impossible. I'm an experienced Exchange engineer and I wouldn't event want to try.
If you put a call into MS this is one of the first things they'd have you do. After that you'd be running one of their SDP (software diagnostic packages).
But before they'd do any serious work, they have you install the latest CU and reboot the server.
Scott C
A cluster node failure would definitely have an impact on performance.
@ScottCha who said running extra command and reading the information from html needed Microsoft internal tools.
i f you dont know about it better to read it .
Extra is not Microsoft internal tool and i am asking to send me the logs information .
For you kind information i am attaching the screenshots here. extra1.PNG extra2.PNG
Thank you all for your input!!!
All is well now, but I would love some more information if you would be so patient with me. I have 20 years Domain experience, so I am not a newb, but little to none on exchange.
When I started, I saw there was a 2 node cluster for exchange in "Failover Cluster Manager" but does not have the resources I would expect with a FC. When the exchange admin came in (who knows barely more then me) he showed me in the web interface where the cluster is at, and indeed under servers > database there were 2 servers. In fact there are 3, with a client front end, and I suspect there may be a fourth for the DB.
So. What keywords should I be studying to look into this.
and, there is no test environment, so what find of commands can I run to look around safely.
Perhaps a better question, what should I NOT do?
I suspect the cluster may be having more issues, but as it is not a standard FC I am not sure how to troubleshoot this. I also have the political complication that I may be "peeing on someone else's tree" if you know what I mean, but I really should be getting up to speed, and my users are more important then someone's (ignorant) pride.
Gerald Connolly
I think you need to establish if there is a resource issue across your cluster, ie checking CPU, Memory, Disk and Network
You also need to establish if something changed from when your cluster worked OK, to its current situation - was a new app installed, a new query/updated query put into production, has the network changed, has there been some rewiring? Have you got some a disk issue, ie a RAIDset that has lost a disk etc etc?
Its unusual for performance issues to just suddenly occur, but occasionally a small incremental change just causes the system to tip over the edge!
There were no changes, it just happened one morning in the middle of the week. Since this is not a normal Windows Failover Cluster, Where would I find the logs?
Scott C
Even something as small as adding a few users can sometime tip the scale.
Now as for learning Exchange....set up your own testing environment.
A fairly decent PC with 16 GB of RAM and an i5 or i7 processor can run vmwareWorkstation so you can set up a lab.
CBT Nuggets, and Pluralsight have good video training.
MS has some really good books as well.
What you should NOT do is learn on a production environment. Even now, I still run commands I'm not sure of in my test lab.
i am ready to help you , first of all login into exchange server open run and type the command extra and follow the wizard and hare me the results .based on this report we can start troubleshooting the actual issue.
Regards