George Brighton
George Brighton
Definitely, some applications are fond of adding a stack trace attribute to error spans. Unfortunately having everything in the trace UI is rather convenient, so clients have a subconscious bias...
Thanks, that all sounds positive. Should the `parquet_row_group_size_bytes` override be reverted when #3038 is merged, or do they work in tandem? I'll empty our bucket when 2.3 arrives - we're...
@joe-elliott That looks great - just waiting for the precompiled binaries to be available in 2.3!
No compactor crashes for the past 2h with v2.3, vParquet3, `parquet_row_group_size_bytes: 10_000_000`, and a new bucket! I'll keep an eye on it. I'll also try to acquire the v2.3 tar.gz...
Just confirming we haven't observed this since v2.3 and vParquet3, though it may be because we haven't run into #2731 since, so there are no corrupt blocks to trip up...
Still looking for a solution to this!
Still looking for a solution to this!
Still looking for a solution to this!
I'm not against this, and would be happy to review a PR!
The bmc package implements IPMI over a network interface. There are commands that may be sent outside a session (i.e. unauthenticated), however these are still UDP packets. I'm not sure...