0 votes

Hi Datalab colleagues,

I am running an image query. I got the following error from time to time. However, when I rerun the script, sometimes the error does not happen again. I wonder why such error happens and disappear sometimes. 

Thanks and best,


/data0/sw/anaconda2/lib/python2.7/site-packages/pyvo/dal/sia.pyc in search(self, pos, size, format, intersect, verbosity, **keywords)
    240         """
    241         return self.create_query(
--> 242             pos, size, format, intersect, verbosity, **keywords).execute()
    244     def create_query(

/data0/sw/anaconda2/lib/python2.7/site-packages/pyvo/dal/sia.pyc in execute(self)
    534            for errors parsing the VOTable response
    535         """
--> 536         return SIAResults(self.execute_votable(), url=self.queryurl)

/data0/sw/anaconda2/lib/python2.7/site-packages/pyvo/dal/query.pyc in execute_votable(self)
    303                 e = self._ex
    304                 raise DALServiceError(
--> 305                     str(e), e.response.status_code, e, self.queryurl)
    306             else:
    307                 raise DALServiceError.from_except(e, self.queryurl)

DALServiceError: 502 Server Error: Proxy Error for url: http://datalab.noao.edu/sia/ls_dr7?SIZE=0.13200190382855773%2C0.131&VERB=2&POS=199.187039%2C7.063755&INTERSECT=OVERLAPS&FORMAT=ALL
by | 389 views

1 Answer

+1 vote
Best answer
I think this error arises because the gateway that handles the requests and responses doesn't respond to a request, and so the service returns an error.  Are you sending many requests through a loop?  If so, it may be that the gateway is overloaded, and slowing down the rate of requests would suppress the error.
by kolsen (2.7k points)
Thanks a lot, Knut. That is true that I am sending the requests through a loop. That is because I need to extract approximately 200-300 image cuts. I hope I did not break down the whole server of Datalab.
No worries!  It might be worth spacing the requests out if you see problems like this.  A queueing system is in planned development, but for now that's the solution, I think.

421 questions

437 answers


640 users

Welcome to Data Lab Help Desk, where you can ask questions and receive answers from other members of the community.