Skip to content
This repository was archived by the owner on Jul 21, 2021. It is now read-only.

Conversation

@ishveda
Copy link

@ishveda ishveda commented Apr 2, 2020

If the node potentially was created by the server, but then client returned an error (e.g. ErrConnectionClose because of networking glitch), the node remains bound to the restored session and prevents acquisition of newer locks, since according to zk recipes for locks, there is a watch created on that node within the following lock, which dead locks further lock acquisition.

it fixes #229

@coveralls
Copy link

coveralls commented Apr 2, 2020

Coverage Status

Coverage increased (+0.9%) to 80.482% when pulling 64ed717 on ishveda:clean-obsolete-nodes into 2cc03de on samuel:master.

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Zookeeper clients contending for distributed lock get into deadlock state during network partition failure

2 participants