DALQueryError: Query Error: Qserv request failed: error:

Support
#1

Hi Folks,

Thank you for your excellent work on DP1, the related paper, and tutorials.

This is also related to DALServiceError DALQueryError with user-uploaded table, running the same query with an uploaded user table for DP1, I get this new error, and I am not sure how to proceed:

---------------------------------------------------------------------------
DALQueryError                             Traceback (most recent call last)
Cell In[54], line 1
----> 1 job.raise_if_error()

File /opt/lsst/software/stack/conda/envs/lsst-scipipe-10.0.0/lib/python3.12/site-packages/pyvo/dal/tap.py:1018, in AsyncTAPJob.raise_if_error(self)
   1016     msg = self._job.errorsummary.message.content
   1017 msg = msg or "<No useful error from server>"
-> 1018 raise DALQueryError("Query Error: " + msg, self.url)

DALQueryError: Query Error: Qserv request failed: error: invalid parameters of the request, ex: CONFIG  addDatabaseFamily the family 'layout_340_3' already exists.

{'method': 'POST', 'timeout_sec': 300, 'url': 'http://172.24.49.51:25081/ingest/database'}
1 Like
#2

Hi @steventgk, thanks for posting this. Could you add what time today you encountered this error and if it was persistent? Sometimes timeouts are temporary.

#3

We see the issue. Trying to fix it without causing an outage first, if that doesn’t work we’ll fix it the hard way.

Thanks for the report, that was a hole in our monitoring. We try to find these things before you do.

#4

(deleting previous reply)

Sorry we thought we had it but we should complete more testing, and we’re keeling over at this point (long day!). Apologies for this, we’ll fix it with fresh brains tomorrow.

#5

Just wanted to say thank you for your efforts on this! Getting DP1 out with all the stretch goals, tutorials and infrastructure is amazing!

1 Like
#6

It is a truth universally acknowledged that developer velocity increases 1% with every compliment, keep it coming :slight_smile:

On a more serious but happy note, we have deployed a no-downtime fix to data.lsst.cloud (pesky double-float conversion mismatch between systems, annoying how these pass their tests until they don’t). Try it and let us know how you got on.

1 Like
#7

@steventgk if you could confirm that the issue is fixed, we can mark this topic as solved?

#8

Hi @MelissaGraham and @frossie, thanks again for getting back to me about this! Unfortunately, I have probably produced a query that returns no results, but it ran to ‘Job Complete’ for a significant time with no errors.

I would likely argue that my query was ambitious, and I should expect a null result; therefore, I am fairly confident that your fix does work, it’s just my bad science!

#9

OK thank you @steventgk ! If you change your mind about the query and think it really should have produced a non-null result, we’d love to see it as an example in a new topic. We’re always learning from user queries (and we like to help!).