-
Notifications
You must be signed in to change notification settings - Fork 290
Labels
bugThis addresses a bugThis addresses a bug
Description
Bareos component version
bareos-dir 23.0.4
bareos-bconsole 23.0.4
Steps to reproduce
- install a new Ubuntu 20.04 system
- set up bareos-repository
- apt update && apt install postgresql bareos-director bareos-bconsole
- run a few concurrent jobs, wait for any one of them to fail with strange catalog errors and strange messages in the PostgreSQL log
Expected results
- PostgreSQL with TLS enabled should work
- Bareos should at least warn me that this will not work properly
- a default installation on Ubuntu should not run into this issue
Actual results
- as described above, jobs fail in strange ways
LOG: SSL error: decryption failed or bad record mac
CONTEXT: COPY batch, line 142906
STATEMENT: COPY batch FROM STDIN
LOG: could not receive data from client: Connection reset by peer
CONTEXT: COPY batch, line 142906
STATEMENT: COPY batch FROM STDIN
LOG: incomplete message from client
CONTEXT: COPY batch, line 142906
STATEMENT: COPY batch FROM STDIN
ERROR: unexpected EOF on client connection with an open transaction
CONTEXT: COPY batch, line 142906
STATEMENT: COPY batch FROM STDIN
LOG: could not send data to client: Connection reset by peer
STATEMENT: COPY batch FROM STDIN
FATAL: terminating connection because protocol synchronization was lost
Environment
- OS: ``
- component:Relevant log output
No response
Relevant traces output
No response
Anything else?
Maybe related to
https://stackoverflow.com/questions/22752521/uwsgi-flask-sqlalchemy-and-postgres-ssl-error-decryption-failed-or-bad-reco according to this post this can happen when multiple threads/processes write to the database over the same connection
The problem is trigger when several jobs runs in parallel and large data are backup
Reactions are currently unavailable
Sub-issues
Metadata
Metadata
Assignees
Labels
bugThis addresses a bugThis addresses a bug