-
-
Notifications
You must be signed in to change notification settings - Fork 507
New issue
Have a question about this project? # for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “#”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? # to your account
Activate persistent_connections_limit generate query error #2352
Comments
Thanks @daikoz We'll look into it. |
I see the master daemon crashed on merging result sets either with or without persistent_connections_limit option |
the issue was fixed at 950e16a you need to update package from the dev repository after CI finished and package will be published to get issue fixed |
Reopening to complete the checklist items |
we need to add to clt test the case for the query into 2 indexes with
should crash daemon or produces internal error message The indexes sizes should be:
The case is the result set from We need:
|
I wrote a Clt-test and tested different variants of
|
Hi, |
6.3.4 is service release and does not include all fixes from the master. I sure this fix along with all other fixes will be in the next major release. You could use package from the dev repository to get this crash fixed. |
…ime when query multiple indexes; fixed sorter schema difference on early exit; fixed #2352
The fix has been included in 6.3.6. |
After test with 6.3.6, I always the issue when I add:
in searchd section |
@daikoz I can't reproduce the issue in 6.3.7:
and so on and no error like:
My config is:
on each node. The mentioned setting is uncommented on both nodes:
The Manticore version is:
Please check if the issue persists in this (or newer) version for you (https://mnt.cr/dev/nightly) and if it does - provide an instruction how to reproduce it. |
I tested on 6.3.7. I can reproduce this issue.
In the scenario, I launch the 2 command while on same server. We can do a zoom, I will share my screen. Same issue on production when I try to activate persistent_connections_limit. |
could you create docker-compose file with indexes and configs we could run locally and catch this issue? |
Bug Description:
I upload data and configuration on s3 (like https://manual.manticoresearch.com/Reporting_bugs#How-do-I-install-debug-symbols?)
In summary, I have 2 servers: SERVER1 SERVER2 with same manticore configuration.
I use a distributed index like:
Scenario:
1/ Configure SERVER1 SERVER2 and modify the ip of agent_persistent
2/ Launch manticore on 2 servers:
searchd --console --config manticore.conf
manticore.conf is provided on s3
3/ on server1, Launch 2 times:
while IFS= read -r ligne; do mariadb -h0 -P9306 -s -N -e "$ligne" 2>&1 >/dev/null || echo $ligne; done < "query.log"
=> All is OK; No errors
4/ Now, uncomment persistent_connections_limit in manticore.conf, restart manticore on 2 servers and launch 2 times on server1:
while IFS= read -r ligne; do mariadb -h0 -P9306 -s -N -e "$ligne" 2>&1 >/dev/null || echo $ligne; done < "query.log"
=> ERRORS:
ERROR 1064 (42000) at line 1: internal error: column 'keywords/keywords' not found in result set schema
Manticore Search Version:
Manticore 6.3.2 c296dc7c8@24062606
Operating System Version:
Debian 12 bookworm uptodate
Have you tried the latest development version?
No
Internal Checklist:
To be completed by the assignee. Check off tasks that have been completed or are not applicable.
The text was updated successfully, but these errors were encountered: