Please stop reporting "database checks are failing" #1166

Activity

1. kesselb

   added the

   Bug

   issue type on Sep 29, 2026
2. kesselb

   removed the

   Bug

   issue type on Sep 29, 2026
3. kesselb

   added

   1. to develop

   on Sep 29, 2026
4. kesselb

   added the

   Bug

   issue type on Sep 29, 2026
5. kesselb

   added

   bug

   on Sep 29, 2026
6. kesselb

   self-assigned this

   on Sep 29, 2026
7. kesselb commented on Sep 29, 2026

   

   kesselb

   on Sep 29, 2026

   Collaborator

   More actions

   Thanks for your report and sorry for the failing setup check.

   I agree that there's probably not much you can do about these as a sysadmin. That said, if you ever spot an inefficient query somewhere, please don't hesitate to open a new issue or send a PR 😉

   Could you share which checks are failing on your instance? On our test instance I see:

   • Alert: Slave_lag
   • Warning: Created_tmp_disk_tables, Join_without_index_rate
   • Notice: Sort_rows, Handler_read_first_rate, Handler_read_rnd_rate, Handler_read_rnd_next_rate

   The Slave_lag alert is a bug on our side. It's already fixed and will ship with the October maintenance release.

   For the rest, I think it makes sense to treat failing notices, like the Sort_rows one you flagged, as acceptable, so they no longer make the setup check fail. Join_without_index_rate also isn't something an admin can really fix, so it should probably be downgraded to a notice as well.
8. W11T commented on Sep 29, 2026

   

   W11T

   on Sep 29, 2026

   Last edited by W11T

   Author

   More actions

   Could you share which checks are failing on your instance?

   On my instance i see

   Performance 5 failing

   High sorted-row rate
   Notice ·
   Sort_rows
   Failing
   Many rows are being sorted server-side.
   Sorted rows average: 53937 per hour.
   Make sure ORDER BY columns are indexed so sorts use indexes instead of merge passes.
   Value
       14.98
   

   Join-without-index rate
   Warning ·
   Join_without_index_rate
   Failing
   Many SELECTs are doing full table scans or joins without indexes.
   Index-less join+scan rate: 813 per hour; aim for less than 1.
   Add indexes for the columns used in JOIN conditions and WHERE filters.
   Value
       0.23
   

   Full-index-scan rate
   Notice ·
   Handler_read_first_rate
   Failing
   Many queries are reading the first index entry, suggesting frequent full-index scans.
   Full-index-scan rate: 363 per hour; aim for less than 1.
   Investigate whether the indexes match query predicates; rewrite queries that scan the whole index.
   Value
       0.10
   

   Fixed-position read rate
   Notice ·
   Handler_read_rnd_rate
   Failing
   Many queries are reading rows by fixed position — usually a sign of full table scans + sorts.
   Fixed-position read rate: 18156 per hour; aim for less than 1.
   Add indexes that match the queries' ORDER BY / JOIN columns.
   Value
       5.04
   

   Sequential-row read rate
   Notice ·
   Handler_read_rnd_next_rate
   Failing
   Many queries are scanning tables sequentially.
   Sequential-row read rate: 321362 per hour; aim for less than 1.
   Add indexes that match common WHERE / JOIN predicates.
   
   Value
       89.27
   

   This is a rather small installation with small tables.
   In such cases it might be reasonable to do table scans etc.
9. kesselb

   closed this as completedin #1167on Sep 29, 2026