Question: Trinity "Exceeded step memory limit" error?
0
gravatar for katharos
12 days ago by
katharos0
katharos0 wrote:

Hi,

I'm trying to assemble a de novo transcriptome using ~270 million paired end reads in Trinity on Galaxy main. The job appears to have finished, although when I look at the details in my history it says

"workdir is /pylon5/mc48nsp/xcgalaxy/main/staging/19349785/working Running Trinity from /local/2956986 slurmstepd: error: Exceeded step memory limit at some point."

However the assembly seems reasonable: ~223,786 transcripts is not surprising given the number of reads that went into it. Did the tool stop prematurely, or is this result ok?

Cheers and thanks, Katie

ADD COMMENTlink modified 7 days ago by dollykc10 • written 12 days ago by katharos0
1
gravatar for dollykc
7 days ago by
dollykc10
dollykc10 wrote:

Sounds like something similar just happened to me.

I tried running Trinity on a dataset with ~38 million reads. After 18 hours, it looked like it finished, everything was green, but the following message was in the history.

Running Trinity from /local/2985876 slurmstepd: error: Exceeded step memory limit at some point. slurmstepd: error: Exceeded job memory limit at some point.

Here's the published history. I emailed it to galaxy-bugs as well.

https://usegalaxy.org/u/coyk/h/unnamed-history

I tried to share this with the galaxy-bugs email instead of publishing the history but I got an error saying this wasn't a valid user.

Coyk

ADD COMMENTlink written 7 days ago by dollykc10
0
gravatar for Jennifer Hillman Jackson
12 days ago by
United States
Jennifer Hillman Jackson25k wrote:

Hello,

This type of error does indicate that the job ran out of memory during execution. You might want to try a rerun to see if the same results are produced. But also please consider sending the data in for review. The Galaxy pre-release for 18.05 is currently installed at Galaxy Main https://usegalaxy.org and we are looking at reported odd behavior. This message would not be expected if the result data is in a success state (green in color).

So, if the dataset is failed (red) a bug report can be sent in. If it is a putative "success" (green) with reported problems in the log, a share link to the history can be sent in instead. Please include the problematic Trinity dataset number and a link to this post for reference. Be sure to leave all input and output datasets active (undelete if necessary).

Support FAQs: https://galaxyproject.org/support/

Thanks! Jen, Galaxy team

ADD COMMENTlink written 12 days ago by Jennifer Hillman Jackson25k
0
gravatar for kmarshall
12 days ago by
kmarshall0
Canada
kmarshall0 wrote:

Thanks for the suggestion Jen!

It did go green, so I was a bit confused! I've just sent an email about it.

Cheers and thanks, Katie

ADD COMMENTlink written 12 days ago by kmarshall0
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 16.09
Traffic: 82 users visited in the last hour