-
Notifications
You must be signed in to change notification settings - Fork 190
Description
This request concerns the 3.2.0 release.
While debugging an issue related to Vault access (we're using HashiCorp Vault with a custom authentication mechanism), we encountered a situation where authentication requests occasionally timed out due to a five-second timeout setting in a client.
We discovered that these timeouts were silently caught in the following code block, with logging only at debug level (which was not active):
Lines 135 to 148 in 15c42fc
| catch (VaultException e) { | |
| if (logger.isDebugEnabled()) { | |
| logger.debug("Unable to determine mount information for [%s]. Returning unavailable MountInfo: %s" | |
| .formatted(path, e.getMessage()), e); | |
| } | |
| return MountInfo.unavailable(); | |
| } | |
| catch (RuntimeException e) { | |
| if (logger.isDebugEnabled()) { | |
| logger.debug("Unable to determine mount information for [%s]. Caching unavailable MountInfo: %s" | |
| .formatted(path, e.getMessage()), e); | |
| } | |
| mountInfo = MountInfo.unavailable(); | |
| } |
As a result, mount information was unavailable, but the code continued attempting to access secrets, which led to "403 Forbidden" errors. It was not immediately clear that the root cause of these access errors was the failure to retrieve mount metadata.
Suggestion: It would be very helpful if such failures were logged at the error or warning level. This would significantly reduce the time required to diagnose similar issues in the future.
Thank you for your work on this project and for considering this request!