[
https://issues.apache.org/jira/browse/KNOX-3431?focusedWorklogId=1039223&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-1039223
]
ASF GitHub Bot logged work on KNOX-3431:
----------------------------------------
Author: ASF GitHub Bot
Created on: 02/Sep/26 14:32
Start Date: 02/Sep/26 14:32
Worklog Time Spent: 10m
Work Description: smolnar82 commented on code in PR #1372:
URL: https://github.com/apache/knox/pull/1372#discussion_r3915215529
##########
gateway-server/src/main/java/org/apache/knox/gateway/services/ldap/interceptor/UserSearchInterceptor.java:
##########
@@ -71,6 +75,7 @@ public Entry lookup(LookupOperationContext ctx) throws
LdapException {
entry = backend.getUser(username, schemaManager);
} catch (Exception e) {
LOG.ldapLookupFailed(ctx.getDn().toString(),e);
+ throw new LdapOperationException(ResultCodeEnum.OTHER,
"Lookup request to backend " + getName() + " failed.", e);
Review Comment:
The only comment I have is if we can somehow get the original backend result
code and use it here other than `OTHER`. I'm not sure about this, so raising it
as a question and would not block the merge of this PR.
Issue Time Tracking
-------------------
Worklog Id: (was: 1039223)
Time Spent: 40m (was: 0.5h)
> LDAP Proxy search should throw exception when any backend fails
> ---------------------------------------------------------------
>
> Key: KNOX-3431
> URL: https://issues.apache.org/jira/browse/KNOX-3431
> Project: Apache Knox
> Issue Type: Improvement
> Components: Server
> Affects Versions: 3.0.0
> Reporter: David Han
> Assignee: David Han
> Priority: Major
> Fix For: 3.1.0
>
> Time Spent: 40m
> Remaining Estimate: 0h
>
> The UserSearchInterceptor currently logs and swallows exceptions when a
> search request to the backend fails. This has the unfortunate effect of
> making it very difficult to tell if there were configuration or network
> errors connecting to the remote backend. Since the exceptions is swallowed, a
> connection error will look like users have been deleted. It would be better
> to raise the exception so it's obvious that there is a problem communicating
> with the remote backend.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
