feat: CON-1346 CON-1407 Don't use rsync to copy the state during recoveries
#3853
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
eichhorl
changed the title
rsync to copy state on the node during recoveries
eichhorl
changed the title
rsync to copy state on the node during recoveriesrsync to copy the state during recoveries
kpop-dfinity
approved these changes
Feb 17, 2025
eichhorl
added
the
CI_ALL_BAZEL_TARGETS
aterga
pushed a commit
that referenced
this pull request
Feb 17, 2025
…coveries (#3853) #2948 introduced a change which avoids using `rsync` during the "upload state" step of recoveries performed on the replica node directly. This PR does the same for the "download state" step by using `cp` instead. Additionally, we add a parameter `--wait-for-cup-node` which is used to specify the IP of the node to be used for polling the current recovery CUP. This is necessary if no IP was specified in either the "download" or the "upload" step. Futhermore, we introduce a new function `confirm_exec_cmd`, which will only execute the given command after explicit user confirmation (similar to how it is already done for `rsync` and `ssh` commands). This allows for easier sanity checks to improve the confidence in the new process. Finally, we add a system test for local node recoveries, which performs a recovery by sending a single `ic-recovery` to a subnet node via `ssh`. Node that, for now, we need to skip the download & merge certification steps, as these require ssh agent forwarding, which seems difficult to do in the test environment. The new system test exercises the recovery variant without version upgrade, failover nodes, or chain key resharing.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
#2948 introduced a change which avoids using
rsyncduring the "upload state" step of recoveries performed on the replica node directly. This PR does the same for the "download state" step by usingcpinstead.Additionally, we add a parameter
--wait-for-cup-nodewhich is used to specify the IP of the node to be used for polling the current recovery CUP. This is necessary if no IP was specified in either the "download" or the "upload" step.Futhermore, we introduce a new function
confirm_exec_cmd, which will only execute the given command after explicit user confirmation (similar to how it is already done forrsyncandsshcommands). This allows for easier sanity checks to improve the confidence in the new process.Finally, we add a system test for local node recoveries, which performs a recovery by sending a single
ic-recoveryto a subnet node viassh. Node that, for now, we need to skip the download & merge certification steps, as these require ssh agent forwarding, which seems difficult to do in the test environment.The new system test exercises the recovery variant without version upgrade, failover nodes, or chain key resharing.