The bor service frequently exit #356
Assignees
Comments
|
The log of heimdall rest: Bor service exit when the log appear |
|
I updated bor with v0.2.14-tmp-span-hotfix, also met the bad case. The bor console executed eth.syncing, the result as following: The bor log: |
|
I'm stuck at |
|
Same thing here. Stuck at |
|
If you are facing this issue on Mainnet, please check your heimdall genesis file. It should be the same as https://github.com/maticnetwork/launch/blob/master/mainnet-v1/sentry/validator/heimdall/config/genesis.json And if it's on Mumbai Testnet, please use bor v0.2.15-beta2: https://forum.polygon.technology/t/v0-2-15-beta2-upgrade-for-bor/2183 |
|
@ssandeep |
|
The error log posted seems to occur when you query the rest-server and it returns a 500 response. AFAIK, things like a retry mechanism and proper error handling is already in place at bor level. Also, other bor nodes aren't encountering this. Can you once try querying you heimdall rest-server on For the query related to bor stuck at a particular block. That was an intermittent issue which caused bor to halt for sometime. But, if you've performed updates to your bor and heimdall nodes, it would work well post that. So, I think the issue might be at the heimdall layer, which also causes the rest-server to send a 500 response. Could you cross check things once at your end with the heimdall node if it got updated correctly or not? |
Heimdall is normal! When the bor synchronous latest block, The bor service will auto exit . |
|
This is my bor start log, It is constantly rebooting. |
|
The bor log: Rest server log: Heimdall log: |
|
Crashes every few hours with a stack like: |
|
@splix the above issue was resolved actually by re-compiling and re-setting bor. What you posted looks quite different from logs and we haven't observed this with any other node as of now. Could you share some more logs of bor running and the flags you're using while running it? Also, what's the behaviour of heimdall at the same moment? |
|
Well, it's actually a much larger stack, I just published last lines here. And I see same stacks in this thread, so I've decided it's the same issue. Maybe I'm wrong and important part was before, will post then on a next crash. I run it with: |
|
now crashed with [part of the log]: |
|
now with: |
|
After another dozen of crashed I was able to gather a whole stack trace: |
|
@manav2401 do you need any additional input from me? I'd really appreciate if you suggest how I can run Bor without crashes. Thanks |
|
@manav2401 with 95% chance it is a data race. The only solution is it cleanup data races, then try again and see if any crushes occurred. 5% stands for "cgo" and "unsafe". |
|
@splix Geth faced with the same a few times, each time it was about OS and libs. Could you check link ethereum/go-ethereum#228 ? |
|
Closing this for now. |
Closed
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
System information
Bor version:
[v0.2.14-tmp-span-hotfix](https://github.com/maticnetwork/bor/releases/tag/v0.2.14-tmp-span-hotfix)Heimdell version:v0.2.7
OS & Version: ubuntu20.04
The text was updated successfully, but these errors were encountered: