-
Notifications
You must be signed in to change notification settings - Fork 6
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
Assertion failed: (pkt.preamble == preamble_cannon_ul && "preamble check"), function rxworker, file rx.c, line 63. #3
Comments
seems to be reproducible here, after only a few seconds I will try with tar at both ends so we can see what file its on when it crashes. Would pcap be helpful? |
when I use tar at both ends, this seems to be stable for much longer, maybe this only occurs with /dev/null sink? |
The preamble check means that the very start of a tcp worker, I'm not sure about the devnull endpoint changing anything
If in doubt, core it and I'll sift through it. |
the log below isn't 100% reliable (tmux scrollback) but is probably still useful:
|
this time with a few ^T sprinkled at both ends:
|
I'll take the pcap now :) , but would you try limiting 'threads 14' on the tx side ? |
this only seems to happen with zfs send, if I use anything else, we're good. how curious. I'll need to wait til tomorrow for a pcap, how can I send it privately? |
I can reproduce this locally; no pcap needed. Stay tuned. |
any progress on this? would be nice to bump the port off 0.89 too |
can you try the current head and see if the buffer changes alter the behaviour. I'd love to crack this, but at the moment it's important that folks send over their bugs. |
still flakey - this is transferring a bit under 100GiB between 2 boxes within the same DC. I initially tried via
|
and another one, much earlier on this time
|
v0.89 built from ports, FreeBSD 12.2R amd64 to FreeBSD 13.0-ALPHA1 amd64.
The text was updated successfully, but these errors were encountered: