-
Notifications
You must be signed in to change notification settings - Fork 902
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
Invalid local_channel_announcement after a successful splice #6572
Comments
Thanks for the report 🙏, this is getting worked on |
Likely related to #6481 |
@ddustin I recklessly added to
to version v23.08.1 and recompiled getting v23.08.1-modded (for both my main node and test node). Added 100k sats to my main node on the channel with my test node. This time the
so no alias. I still got a
However, this time it was enough to restart my test node and everything was normal at my main node:
So after the restart of my test node the splicing worked perfectly! Of course, normally the other node (the test node) isn't yours, so a restart won't happen. |
Thanks for your reckless testing! It's possible that the (future) release could touch your issue, as a good handful of fixes are in #6677 but in either case a regression test for this sparse |
I was able to recreate the Likely this would already be caught by |
@ddustin I first did another trick in the software to get master going again but then I ran the query command manually directly on the database as you suggested (thanks!) and got master also running again this way. I get the impression the channel is stuck in "CHANNELD_AWAITING_SPLICE" because more than 30 confirmations have already passed. This is from
I hope this isn't because of the (small) software trick I did first in runes but I don't think so. I'll be sleeping and busy but hope this is useful :-) |
Problem is I can connect with my test node to other nodes but not to my main node as I get:
|
@ddustin I can't connect with my test node to other nodes without crashing so I think my test node database is corrupt. I will close and remove my test node and start from scratch. I won't be able to test anything for some time, so just ignore what I wrote lately. At least it isn't my main node :-) |
@ddustin think I found the source of all my problems in the source code. See my pull request :-) |
@ddustin After adding the extra ?, everything ran perfectly. I've never seen anything like it, what a difference. Splicing worked, including |
That's awesome! Thanks for figuring that out! 🔥 |
* Regression test added for Issue ElementsProject#6572 (issuecomment-1730808863) w/stuck HTLC * `check_balance` adjusted to calculate pending HTLCs explicitly * Test confirmed to fail prior to PR ElementsProject#6713 ChangeLog-Fixed: Issue splicing with pending / stuck HTLCs fixed.
* Regression test added for Issue ElementsProject#6572 (issuecomment-1730808863) w/stuck HTLC * `check_balance` adjusted to calculate pending HTLCs explicitly * Test confirmed to fail prior to PR ElementsProject#6713 ChangeLog-Fixed: Issue splicing with pending / stuck HTLCs fixed.
* Regression test added for Issue ElementsProject#6572 (issuecomment-1730808863) w/stuck HTLC * `check_balance` adjusted to calculate pending HTLCs explicitly * Test confirmed to fail prior to PR ElementsProject#6713 ChangeLog-Fixed: Issue splicing with pending / stuck HTLCs fixed.
* Regression test added for Issue ElementsProject#6572 (issuecomment-1730808863) w/stuck HTLC * `check_balance` adjusted to calculate pending HTLCs explicitly * Test confirmed to fail prior to PR ElementsProject#6713 ChangeLog-Fixed: Issue splicing with pending / stuck HTLCs fixed.
* Regression test added for Issue ElementsProject#6572 (issuecomment-1730808863) w/stuck HTLC * `check_balance` adjusted to calculate pending HTLCs explicitly * Test confirmed to fail prior to PR ElementsProject#6713 ChangeLog-Fixed: Issue splicing with pending / stuck HTLCs fixed.
* Regression test added for Issue ElementsProject#6572 (issuecomment-1730808863) w/stuck HTLC * `check_balance` adjusted to calculate pending HTLCs explicitly * Test confirmed to fail prior to PR ElementsProject#6713 ChangeLog-Fixed: Issue splicing with pending / stuck HTLCs fixed.
Issue and Steps to Reproduce
I promised to retry a splice after my previous issue was fixed. All went fine and 100 ksats were added to my main node, except that my test node (
02aecde3b3dde450889755d4dbce538e6b9e9dfa7702d14b5bc5c4379e1b330392
) was no longer visible inlistnodes
of my main node (as was briefly mentioned in my previous issue), although it was visible inlistpeerchannels
of my main node. I uselistnodes
to get aliases. Restarted my test node and this time it didn't crash anymore! After the restart of my test node, it was visible in thelistnodes
of my main node but without the alias and other information:while others show more info, for example:
Note that my main node is perfectly visible in the
listnodes
of my test node. Restarted my main node but that didn't help, the alias of my test node is still not visible in thelistnodes
of my main node.Here is the output of my main node (see UNUSUAL and BROKEN):
EDIT:
1: stopped my test node, removed the test node gossip_store and restarted the test node -> alias still not visible at my main node
2: stopped my test node, deleted payments and invoices from the test node lightningd.sqlite3 -> alias almost immediately visible again at my main node (
sqlite3 -header -line lightningd.sqlite3 'DELETE FROM payments'; sqlite3 -header -line lightningd.sqlite3 'DELETE FROM invoices'
)getinfo
outputBoth nodes are: "version": "v23.08rc2-17-gc67f1f9",
The text was updated successfully, but these errors were encountered: