Parseable Service goes into panic randomly

[imagdy83]

Dashboard doesn't show any data journalctl -u parseable shows the following

parseable[3597327]: thread '' panicked at 'called Option::unwrap() on a None value', server/src/storage.rs:352:13

parseable[3597327]: thread '' panicked at 'called Result::unwrap() on an Err value: PoisonError { .. }', server/src/storage.rs:138:61

[nitisht]

Thanks @imagdy83 we'll look into this quickly. What is the server version here?

[imagdy83]

Current Version: 0.0.7

[trueleo]

@imagdy83 Hello, thanks for reaching out. I will check why it happened but meanwhile can you restart parseable and wait for minute or two then check following

• there should be no data in the data directory
• any data that failed to upload previously should upload to s3

[trueleo]

@imagdy83 also try to use latest release after you have let your existing data upload to s3. We have released v0.1 https://github.com/parseablehq/parseable/releases/tag/v0.1.0

[imagdy83]

@trueleo

• What do you mean there shouldn't ne data in the data directory? there is a folder for each stream we've created
• Restarted parseable and waiting for it to panic again

[imagdy83]

@trueleo Can you provide the steps to update from 0.0.7 to 0.1.0

[trueleo]

@imagdy83

What do you mean there shouldn't ne data in the data directory? there is a folder for each stream we've created

there should be a directory called data ... it is created in the current directory relative to where parseable executable is

[trueleo]

@trueleo Can you provide the steps to update from 0.0.7 to 0.1.0

If you check 0.1 release notes we do not support upgrade from <0.0.8 to 0.1. If you cannot do a clean setup and want to migrate we are happy to help on Slack. https://launchpass.com/parseable

I will ping you here if I find it possible to reuse the bucket you created on 0.0.7 for 0.1

[imagdy83]

the service file is corrupt can you provide a service file for linux please

[imagdy83]

parseable.service: Starting requested but asserts

[nitisht]

The command to start the server has changed (need to add s3-store after parseable), this is the new service file https://github.com/parseablehq/parseable/blob/main/systemd/parseable.s3.service -- for v0.1.0

[trueleo]

@imagdy83 If you want to use your existing bucket (which was created on v0.0.7) you will have to rename a file on s3. stream_name/.parseable.json to stream_name/.stream.json . You can do this through s3 move command or by downloading file and re uploading under the new name.

[imagdy83]

So far it has not paniced, I'll keep you posted thanks

[imagdy83]

Good Morning we've hit a ton of errors looking like these parseable[3612606]: thread 'actix-rt|system:0|arbiter:0' pani

parseable[3612606]: thread 'actix-rt|system:0|arbiter:0' panicked at 'no poisoning: PoisonError { .. }', server/src/query/table_provider.rs:57:70

[trueleo]

@imagdy83 Can you tell me about you setup. How did you set parseable up and what your typical usecase is like.

[imagdy83]

• We've got several sources with logs
• Each of the sources forwards their logs to a syslog-ng instance that's deployed on the same server that parseable is installed on
• Syslog-ng has one of it's destinations as the parseable server and there is another log forwarder on a different source that forwards directly to parseable
• logs naturally get uploaded to s3 in the process

[nitisht]

@imagdy83 would you mind getting on a short call to address this? The use case is exactly the kind we're looking at and we'd love to fix this asap for you.

Also have you upgraded to v0.1.0 already?

[imagdy83]

Sounds good, don't mind yes we're at v0.1.0 at the moment

[nitisht]

okay, I am at nitish@parseable.io , please drop a note and we'll schedule the call

[imagdy83]

sent you an email