Hi everybody!
I’m using move_chunk function for moving chunk from one hypertable to another. Chunk contains 300 millions rows. After a while I get an error: server closed the connection unexpectedly. This probably means the server terminated abnormally before or while processing the request.
What is the reason of it ?
In the postgresql.conf I have: listen_addresses = ‘*’, so I don’t know what to do.
Hi @Wilgo thanks for the question
Are you seeing any error messages in your Postgres or server logs that coincide with this?
Also, just to confirm you are using multi-node? Is it on all hypertables ? And could you share the version of TimescaleDB and the OS environment? Finally, is this a long running database or new? And did you upgrade recently?
So sorry – a lot of questions! But any information you can share may help someone diagnose the issue. Thanks!
Hi, @LorraineP !
Thanks for the answer!
In my postgres log I see this:
[2458] LOG: background worker “Reorder Background Job” (PID 5009) was terminated by signal 11: Segmentation fault
[2458] LOG: terminating any other active server processes
[4908] WARNING: terminating connection because of crash of another server process
[4908] DETAIL: The postmaster has commanded this server process to roll back the current transaction and exit, because another server process exited abnormally and possibly corrupted shared memory.
[4908] HINT: In a moment you should be able to reconnect to the database and repeat your command.
[4427] WARNING: terminating connection because of crash of another server process
[2458] LOG: all server processes terminated; reinitializing
[5090] LOG: database system was interrupted; last known up at 2022-05-16 11:21:45
[5090] LOG: database system was not properly shut down; automatic recovery in progress
[5090] LOG: redo starts at D1C8/233C4060
[5090] LOG: redo done at D1C8/233EDFA0
I chose some strings which should be useful.
Sorry, I don’t understand how to answer the question about the multi-node, because I “inherited” this entire database :((
In this base are five hypertables and many chunks.
Version of TimescaleDB is 1.7.1, OS - Red Hat Enterprice, Linux 3.10.0-1127.
This is a long running database and I haven’t done any updates lately.
Hello again @Wilgo and I’m so sorry for the delay in responding.
Talking with the Timescale team they feel your first step really needs to be to bring your version of TimescaleDB up to date as many things have been fixed since that version. If you do that you may find it resolves the issue. The updating guide starts here (but be aware of special steps to move between major versions https://docs.timescale.com/timescaledb/latest/how-to-guides/update-timescaledb/)
Would you be able to take a look at that maybe? 