Bug 1295499 - Attempts to update a specific search query results in infinite "Please stand by"
Attempts to update a specific search query results in infinite "Please stand by"
Product: Bugzilla
Classification: Community
Component: Query/Bug List (Show other bugs)
Unspecified Unspecified
medium Severity medium (vote)
: ---
: ---
Assigned To: PnT DevOps Devs
Depends On:
  Show dependency treegraph
Reported: 2016-01-04 12:33 EST by Tom Lavigne
Modified: 2017-02-01 06:41 EST (History)
3 users (show)

See Also:
Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of:
Last Closed: 2017-02-01 06:25:19 EST
Type: Bug
Regression: ---
Mount Type: ---
Documentation: ---
Verified Versions:
Category: ---
oVirt Team: ---
RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: ---

Attachments (Terms of Use)

External Trackers
Tracker ID Priority Status Summary Last Updated
Red Hat Bugzilla 1260009 None None None 2016-01-04 18:19 EST

  None (edit)
Description Tom Lavigne 2016-01-04 12:33:47 EST
Description of problem:

Attempts to edit a specific search query results in a "Please stand by" that never ends.  Infinite loop?

Version-Release number of selected component (if applicable):

How reproducible:

Always with this one specific query.   I have tried multiple edits to the same query without being able to workaround the problem.

Steps to Reproduce:
1.  Use this URL, which is a relatively complex search query, which DOES finish:  https://url.corp.redhat.com/42c17a4
3. Remove the first line in the query, the "blocks" criteria.
4. Search

Actual results:

"Please stand by" forever

Expected results:

A list of bugs resulting from the updated search.

Additional info:
Comment 1 Jeff Fearn 2016-01-04 18:12:26 EST
I tested this as instructed, and the results are as stated.

I then added:

Classification: Red Hat
Product: Red Hat Enterprise Linux 6

and reran the search, 19 bugs found in ~5 seconds.

The question now is, why doesn't the search timeout like it's supposed to?
Comment 2 Jeff Fearn 2016-01-04 18:18:44 EST
FYI This relates to the way Postgres handles the search and isn't a code loop, the DB is taking a long time to check the "bug_id NOT IN(~1 million ID's)". Setting the product reduces the number of IDs by a factor or two and thus computes reasonably fast.
Comment 3 Jeff Fearn 2017-02-01 06:25:19 EST
The proxy was tweaked recently and this query now correctly dies with the "Your query was taking longer than 120 seconds and had to be killed.  Please back up and narrow your search criteria ..." message.

Given that taking the advice in the message gets a search result I'm going to close this. Jump on Bug 1260009 if you want to pursue improving the search generation.
Comment 4 Tom Lavigne 2017-02-01 06:41:02 EST
Agreed, sorry should have closed the bug before.  I left it open in case you wanted to improve the messaging, which it looks like you have done according to comment #3.  Thanks!

Note You need to log in before you can comment on or make changes to this bug.