-
Notifications
You must be signed in to change notification settings - Fork 20.9k
panic during normal operation #1955
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
Comments
Do you recall a sudden shutdown (e.g. power loss) last time you ran |
I don't remember. It's unlikely, but possible. geth synced 80k blocks in about an hour before crashing:
|
i had the same panic, comment here: #1962 (comment) (related tickets?) |
@donSchoe The two (Actually three) issues are unrelated, please open a new issue for yours. |
@dooglus @obscuren I've been trying to wrap my head around this one (line producing the crash 565d9f2#diff-1549393320d640c34a2a9e27e4194adfL60), but I don't really see a possibility for the crashing code to produce the panic on itself. My bet is that there was a deeper data race in there, and while the hex decoding was running another process swapped out the underlying slice. That would explain where the nil dereference comes from. As @fjl quite significantly changed the trie since then, I'm not sure this issue is relevant any more, but perhaps @fjl can check if there was a possible data race in the old code and if that is still present or not in the new version. |
The older trie code didn't handle invalid nodes very well. I'd rule out a race issue because the older trie also had |
I'll close his for now since the new code is very different and this issue hasn't popped up again. |
* fix: use the top root hash for rewinding under path schema * feat: add async flush nodebuffer in path schema * chore: add prun-block param suffix check * fix: code review comments
I ran 'geth console' for the first time in a couple of months. It was syncing nicely, but taking a long time. I left it syncing overnight.
When I woke up it had crashed:
It reports this version string:
I realise I'm running an old version and that this bug may already have been fixed, but thought I should report it anyway since I can't find mention of it in the issues list.
The text was updated successfully, but these errors were encountered: