https://signoz.io logo
Join Slack
Powered by
This message was deleted.
# support
s
This message was deleted.
p
@Vishal Sharma any insights here?
a
I ended up nuking the clickhouse instance to resize it to a larger EC2 instance type and the SigNoz UI seems to have remembered it was still pending on changing retention to 30 days and I can’t seem to set it to 30. Is there some kind of timer where it keeps retrying in the signoz app? Can i somehow wipe or reset this without nuking the entire signoz DB?
I ended up nuking the SigNoz services + filesystem on EFS to start fresh. Hoping this gets looked at as when this actually gets used in PROD would not want to send an invite link to everyone again
v
@Alexei Zenin TTL is an async operation which is why it’s taking so long. If you want to make the changes faster then you can manually connect to clickhouse and then reduce 
merge_with_ttl_timeout
to 10mins, this has performance impact as TTL will be executed frequently. Performance impact of this can be reduced by enabling 
ttl_only_drop_parts
. Also, if your table is not too big, you could force it using the 
OPTIMIZE TABLE [db.]table FINAL
statement. However, for large tables, it is not recommended.
Command to set 
merge_with_ttl_timeout
to 10mins:
Copy code
SET merge_with_ttl_timeout = 600
Command to enable 
ttl_only_drop_parts
Copy code
SET ttl_only_drop_parts = 1
If you want to run 
OPTIMIZE TABLE [db.]table FINAL
for all the trace tables then run below queries one by one: 1. 
OPTIMIZE TABLE signoz_traces.signoz_error_index_v2 FINAL
2. 
OPTIMIZE TABLE signoz_traces.usage_explorer FINAL
3. 
OPTIMIZE TABLE signoz_traces.dependency_graph_minutes FINAL
4. 
OPTIMIZE TABLE signoz_traces.signoz_spans FINAL
5. 
OPTIMIZE TABLE signoz_traces.signoz_index_v2 FINAL
6. 
OPTIMIZE TABLE signoz_traces.durationSort FINAL
59 Views
Open in Slack
PreviousNext
Powered by