cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

Getting lot of early discarded purepaths

tarun_agastya
Organizer

Hi,

we are seeing early discarded pure paths.can someone help us what might be the reason for this.

early-discarded.png

Thanks,

12 REPLIES 12

JamesKitson
Dynatrace Leader
Dynatrace Leader

Here's a short description of the meaning:

Early Discarded Purepath:
This means Purepaths are dropped, but have made it to the dynaTrace Server. For example, they can be dropped during correlation, before or during RealTimeAnalysis, or due to low memory (and we are clearing out buffers).

What version are you on?

Perhaps it is related to the particular version (although it seems when this is the case it is more pronounced) or some issues with the Dynatrace server itself, maybe a screenshot that includes other server metrics around the time you are seeing these early discarded PurePaths would help.

If this gets significant I would recommend opening a support ticket,

James

david_c_johnson
Advisor

We were having the same issue. I increased the memory allocation of the server and it resolved. see in my attached image. we updated from 6.1 to 6.3 around the 14th and the memory setting was changed around the 16th.(reduced) after that nothing but issues with early discarded. on the 1st we increased the mem setting. Green pastures after that.

Thanks for the update. Did you upgrade to 6.3 or to the latest 6.3.x?

Hi,

like @Benjamin Wrightson mentioned in his comment: we have a customer with the same problems: from time to time we have many early discarded purepath - we can only get back to the normal monitoring state with restarting dynatrace services.

We have 6.3.3 running and did no changes during upgrade.

Best regards

Friederike

benjamin_wright
Dynatrace Pro
Dynatrace Pro

Other customers have reported similar issues here.

andreas_grabner
Dynatrace Guru
Dynatrace Guru

I suggest to reach out to our support team by opening a ticket. if this is an issue that more people are seeing I am sure our support team is already aware of it and can provide a solution. would be great if you could update the forum entries!

thun
Advisor

Support told us to just go ahead and install the Fixpack 4. Just for others who might have the same issue.

Hi Friederike, did the fixpack fix the issue?

Babar_Qayyum
Leader

Hello @James K.

Need your advice/assistance/best practices to avoid the situation and please also correct my understanding which I am trying to make my point in the below analysis.

I did some analysis to correlate the reasons behind Skipped PurePaths and the Queued UEM Events. My understanding that whenever the correlation queues will be high (what could be the reason behind enormous Queued Events suddenly?) then the number of skipped PurePaths will be increased.

Below screenshot is an example which is representing the same time-frame.

One more observation about the Dropped/Skipped User Actions which are the same as Early Discarded and Skipped Non-Analyzed PurePaths.

Below screenshot is an example which is representing the same time-frame.

Note: Dynatrace Sever, Performance Warehouse and Session Storage are working accordingly or within the recommended thresholds as per the sizing.

Regards,

Babar

I'm not familiar with anything that could cause that. You may want to check with support if you cannot correlate that with any events at that time (something like a server or collector issue that maybe caused things to queue up and the require it to start skipping analyzing data after functionality is restored).

Hello James,

Sure. I will open the case with the Support.

Thank you for always being available.

Regards,

Babat

christian_wegha
Advisor

Also having this problem on version 7.2.5 ....

Is it recommended here to upgrade to latest DynaTrace Appmon Version (V7.2.12)?