Every now and then I start getting "Unexpected. Please try again." for ANY query i try. It usually gets better a bit later but the rate at which it appears is very worrying.
What can I do when this happens?
here are some job ids:
job_648552a52f3046c5b2df9300a31d4693
job_4af4184725974c3fb38e0ded96b776c9
If you get this response, please let us know because it indicates a bug in BigQuery.
Looking at the jobs in question, this is a race condition where we mark the import complete before it has completely replicated to all datacenters, and the query hits a different datacenter. I've filed a bug and we're working on a fix.
Related
I am currently not able to save query results to a new table in BigQuery using the BigQuery console. I was able to do this 2 weeks ago, but now it doesn't work for me or anyone in my organization. I don't get an error; the 'loading' wheel simply keeps indicates that it's loading, and it eventually times out.
Has anyone experienced this? I thought it was a general BigQuery issue, but there's no evidence of others complaining, or a general bug.
[Image of what occurs when I try to export results to a BigQuery table1
Me too. I am unable to export query results to BQ table since yesterday. I thought it is just me, now I know everyone is affected. Think it is bug that needs to be fixed, quick!!!
LATEST: Go to Query Settings and change processing location to somewhere else and you should be able to save the query results. I think default is USA, and I changed to Singapore. Try a few locations and see which one works for you.
I have the same problem. I'm gcp project owner and my Bigquery quotas are almost in 0% I noticed this problem since the day before yesterday and it's very annoying. Currently I have to export results into a file (csv or json) in order to import them later from this same file. I hope Google fix this bug soon.
I am having the following problem: I have set up Google Ads data transfer to Bigquery, it used to work fine for about a month, then I got some AuthorizationError.USER_PERMISSION_DENIED error.
Now, I have seen here that potential cause might be
"The link between the manager and client account was removed."
but that does not explain why transfer schedule had successful runs after first (few) errors, and then stopped working again. Can anyone shed some light on this, please? How can I fix the problem?
Here are screenshots of run history, more run history and transfer configuration
P.S. Start date was changed to "December 1" instead of early November after I rescheduled time.
All my datasheets, tables, and ALL items inside BQ are un EU. When I try to do a View->to->Table 15 min scheduled query I get an error regarding my location, which is incorrect, because all, source and destiny are both on EU...
Anyone knows why?
There is a transient known issue matching your situation, GCP support team needs more time for troubleshooting. There may be a potential issue in the UI. I would ask you to try the following steps:
Firstly, try to make the same operation in Chrome's incognito mode.
Another possible workaround is trying to follow this official guide using a different approach than the UI (CLI for instance).
I hope it helps.
I observed something very strange today when trying to steam records into bigquery table , sometimes after the successful stream, it shows all the records being steamed into, something it only shows part of it? What I did was I deleted the table, and recreated it. Has anyone encountered any scenario like this? I am seriously concerned.
Many thanks.
Regards,
I've experienced a similar issue after deleting and recreating the table in a short time span, which is part of our e2e testing plan. As long as you do not delete/recreate your table streaming API works great. In our case workaround was to customize streaming table suffix for e2e execution only.
I am not sure it this was addressed or not, but I would expect constant improvement.
I've also created a test project reproducing the issue and shared it with BigQuery team.
Today, after many successful loads into a BigQuery table, received this error message:
tableUnavailable
Something went wrong with the table you queried. Contact the table owner for assistance
I do not see this error in the error table: https://cloud.google.com/bigquery/troubleshooting-errors#errortable
What conditions could cause this error? Other load jobs, using the same code and in same dataset, do not display this error
What causes a "tableUnavailable" message?
There are two cases that I can think of:
First, this error can be returned for queries over a (small) set of tables that BigQuery exposes access to, but are not directly managed by the BigQuery team itself. You can consider these equivalent to "internalError" from a troubleshooting perspective.
These data sources are typically accessible to GCP customers that have specific relationships with Google product teams exposing their data in BigQuery.
We expose these under a different error code since you will resolve the issue more quickly by contacting the group that granted you access to their data. Going through BigQuery customer support to get this resolved will work too, it'll just take a little longer.
Second, you encountered this through a load job so this is clearly not the case above! We are testing a new load implementation that is faster than the current implementation, and I suspect some errors are mapped slightly differently now.
In this case, I suspect you encountered a "backendError" and should try the operation again. If you can give us a project_id:job_id of a job that hit this problem, we can verify this and make sure the error mapping is more consistent.
Thank you!