trouble uploading target/query files

[jasonwjohns]

Describe the bug
In the 'Launch map analysis' feature, using the 'New alignment' function, I'm having trouble uploading target and/or query files. I submit the job and the files start to upload, then I get an error message saying that either the target or query file could not be uploaded, and to contact to report the bug. The issue has been going on for a few days. In that time, a couple of alignments have successfully loaded out of a couple dozen attempts, but it's inconsistent. I've been making attempts mostly with files that I have previously successfully aligned.

To Reproduce
Steps to reproduce the behavior:

1. Go to Run
2. Load target and query files
3. Click 'Submit'
4. See error

Expected behavior
I expected the files to upload, the alignment to run and to be able to view the results, as usual.

Screenshots

Computer (please complete the following information):

• OS: mac
• Browser: chrome
• Version: 120.0.6099.129

Additional context
Add any other context about the problem here.

[chklopp]

Hi,

In our database I see all these jobs (job IDs) which were ran successfully :
exi_target_coe_query
exi_target_kan_query
coe_target_kan_query
exi_target_coe_query_2
exi_target_kan_query_2
coe_target_kan_query_2
kan_target_exi_query
coe_target_exi_query
kan_target_coe_query
coe_target_exi12_query
Shh6C_20240103220832
zwr77_20240104010108
l9U1A_20240104190219
Some completing yesterday.

Could you give me the ID and time of a failed upload?

Cheers,

[jasonwjohns]

Hi Christophe,

Thank you for looking into my inquiry. One of the jobs I've been trying to run for the past couple days is "kan_target_exi12_query". I've tried it several times, but the most recent attempt was at about 8:30 PST this morning (1/5/23). Below are screenshots of the two different error messages I get. I haven't been able to identify a pattern associated with whether I get one message or the other.

Over the past few days, I've probably made a couple/few dozen attempts at doing alignments, and most of them have failed. As you said, a few of the jobs you listed completed yesterday, but those represent a small portion of the ones I have attempted. The jobs with the default names are ones that I submitted as a test, as they have worked previously. Most of my successful jobs are from weeks ago.

On an unrelated note (I think), none of the results are ever emailed to me, as stated on the website. They used to be emailed to me in the past, but are no longer. I've tried using different email addresses and have checked Spam folders. It hasn't really been an issue, because I can just access the cached results directly from the site, but I thought I would let you know in case you wanted to look into it.

Thank you for your help!
Jason

[pbordron]

I looked into the logs and don't find trace of a job named kan_target_exi12_query, because until the upload don't finish, the job is not created.

However we had many session timeout errors. This must append when the upload session timeout or when the upload ping failed. As the flask server we use is not asynchronous, there is some cases where it can timeout if there is too much huge upload simultaneously.
It is something we already try to mitigate, but it need further developments to solve definitely this issue.

I have restart the service. Can you try again?

Regards

[jasonwjohns]

Hi Philippe,

Thank you for your explanation. I just tried again and it seems to have timed out again. The two genomes I'm trying to upload are ~285 Mo each and I have Minimap selected. The other day I tried to upload files with Mashmap and it also timed out.

Jason

[pbordron]

Look like it is something else. I tried to run a job, but my upload doesn't start even when my upload session is active... Must investigate that with the network team.

[jasonwjohns]

On my most recent attempt, both the target and query files loaded and the alignment worked fine. Must have been a lucky submission?

[rotifergirl]

I am having the same issue and have not managed to have a successful attempt at all today

[celine304]

I'm also having the same issue trying to upload the target file

[pbordron]

There are some network problem on the inrae infrastructure that cause upload to fail

[pbordron]

Solved

[jasonwjohns]

I'm not seeing that the issue was resolved, as the jobs I've tried to submit have continued to fail ("zostera_target_phyllo_query"). Is it an issue on my end?

…

On Tue, Jan 16, 2024 at 1:26 AM Philippe Bordron ***@***.***> wrote: Closed #117 <#117> as completed. — Reply to this email directly, view it on GitHub <#117 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AIYTVHHCI3LLTSAHANBFOTLYOZBVFAVCNFSM6AAAAABBNFU6WGVHI2DSMVQWIX3LMV45UABCJFZXG5LFIV3GK3TUJZXXI2LGNFRWC5DJN5XDWMJRGQ4TMMRWHAZTONI> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[pbordron]

We will change the way dgenies is deployed during the cluster maintenance.