refactor(pinmfs): log error if pre-existing pin failed #8056
Conversation
This cleans up debug logging and ensures we use %q for data read from remote services. We log pre-existing pins for MFS root and print ERROR if pin for current root has status "failed", otherwise, we log to debug.
lidel
force-pushed
the
log_ERROR_if_pin_failed
branch
from
78d64b4
to
3b4c615
Compare
There was a problem hiding this comment.
Thank you, @BogdanStirbat, LGTM (tweaked messages a bit, so things look more uniform now)
@Stebalien / @aschmahmann this should be ready for final review.
FYSA this PR does not change any functionality, only adds/tweaks logs produced on ipfs log level remotepinning/mfs on error and debug levels
| if ps.GetPin().GetCid() == cid && ps.GetStatus() == pinclient.StatusFailed { | ||
| mfslog.Errorf(pinStatusMsg, svcName, pinclient.StatusFailed, cid, existingRequestID) | ||
| } else { | ||
| mfslog.Debugf(pinStatusMsg, svcName, ps.GetStatus(), ps.GetPin().GetCid(), existingRequestID) | ||
| } |
There was a problem hiding this comment.
This is the only real change in this PR, everything else is just my opportunistic cleanup of already existing debug logs :)
👉 error log level is used only when failure is for the current MFS root CID, everything else is logged to debug:
There was a problem hiding this comment.
Isn't this a bit awkward since if your node is offline and you start mutating MFS your log will fill up with errors?
There was a problem hiding this comment.
No, pinning is lazy atm, so service will return queued status and then try to pin it for some time (eg. Pinata won't return failed until N hours passed without any new bytes found).
In practice, the ERROR added in this PR is only printed once, when you restart node.
Main value here is improved logs when ipfs log level remotepinning/mfs debug is set.
lidel
changed the title
BigLep
added
kind/enhancement
* pinmfs: log error if pin failed * refactor: log pre-existing remote mfs pins This cleans up debug logging and ensures we use %q for data read from remote services. We log pre-existing pins for MFS root and print ERROR if pin for current root has status "failed", otherwise, we log to debug. Co-authored-by: Marcin Rataj <[email protected]> (cherry picked from commit 075cc8f)
* pinmfs: log error if pin failed * refactor: log pre-existing remote mfs pins This cleans up debug logging and ensures we use %q for data read from remote services. We log pre-existing pins for MFS root and print ERROR if pin for current root has status "failed", otherwise, we log to debug. Co-authored-by: Marcin Rataj <[email protected]> (cherry picked from commit 075cc8f)
Add an error log message to
remotepinning/mfs, in case a pin request failed.