-
Notifications
You must be signed in to change notification settings - Fork 1.4k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Ord index terminating every few thousand blocks #813
Comments
This data is from running 325201f In total, required 98 runs of the index command. Total time to run was from 2022-11-15T21:51:35Z to 2022-11-18T15:32:06Z so roughly 64 hours. Time and block height of stoppages attached. |
Running with latest master from today - interestingly the first block heights where it terminates are identical, but not the third one:
|
Second run with latest master from Friday (fa79995) finished. Command I am using:
In total, 86 runs of the index command. Total time to run was from 2022-11-18T20:37:53Z to 2022-11-20T01:38:11Z so roughly 31 hours. Time and block height of stoppages attached. |
It's a JSON-RPC error, somehow our retry does not work:
|
Turns out this is just https://github.com/casey/ord/issues/747 and we should probably resurrect https://github.com/casey/ord/pull/748 |
This is no longer a problem, retrying in #820 means that we continue even if we get an error. However, this is still happening, and the pattern is so regular that I think we must be doing something weird:
It starts around block 365034, and from there we get an error every time |
This is low priority, since it's not actually causing any problems, so pushing it back to the production inscriptions milestone. |
Maybe, not so much of a mystery. The MPSC channel that the background block fetcher uses to send blocks to the main thread can queue up 32 blocks.
My guess is that this is related to the 15 second timeout in |
FYI I'm seeing this on my linux system on latest master. |
I suspect latest master is particularly bad here, because of #1166, so flushes are taking a lot longer, so we're more likely to hit the timeout. |
I face this issue with Below is sample logs:
Then I broke and run the server, yet again. |
My mac with Xeon processor and ECC RAM crashed while indexing blocks and being stuck at 580000 for about 18 minutes. Below is the panic log:
Until that point, the following blocks hanged for less than 30 seconds each: After the crash, the indexing started from ~577K again. I've ord 0.4.2 running on macOS Ventura. |
i am getting similar issues on Mac OSX 12.6.3, Mac Pro 7,1 tower (2019 cheese grater), plenty of horsepower & 250GB SSD internal space left. When using I have been trying for a week, incredibley frustrating. "ord index" restarts from 6555000 and continues to hang at 659956. repeats the same thing, have repeated 20+ times... Have tried disconnect bitcoind and restarting, also bitcoin reindex (took forever but work successfully), running out of options... I have no idea what to do here... incredibly frustrating, appreciate any advice... |
Crashed with a panic again. 0.4.2 won't fool me again. Moving on to build from the master branch. |
10f8469 is the first bad commit |
This definitely seems like a bad bisect. That commit is super harmless. |
Yes, that's a bad bisect. I meant to write that to https://github.com/casey/ord/issues/1467. Please ignore. |
seeing similar behavior on AWS hosted windows machine. was smooth sailing until block 335,000. after that getting the below error consistently every 5000 blocks
|
Index still running 13 hours later with the same behavior (error every 5000 blocks) on the ☝🏾 system. I started this process on a locally hosted machine and got the error much earlier (block 29400, block 122629, block 129636). No discernible pattern on when the error is occurring on this system. Hope it helps! |
Should be fixed by https://github.com/casey/ord/pull/1516. |
Haven't seen this issue come up since February. |
Still seeing this issue: |
I am suffering it now. Every 5000 blocks , it comes up with rocks error. |
I'll log a new issue and ref this one then. @DrJingLee |
maybe you should try to download here I stopped reindex for the last whole week and just use the new version of index.redb. |
Reported by @virtuallythere:
The text was updated successfully, but these errors were encountered: