| Summary: | Improve a confusing log message related to LimitRequestLine | ||
|---|---|---|---|
| Product: | Apache httpd-2 | Reporter: | Jackie Zhang <jackie.qq.zhang> |
| Component: | Core | Assignee: | Apache HTTPD Bugs Mailing List <bugs> |
| Status: | RESOLVED FIXED | ||
| Severity: | major | CC: | jackie.qq.zhang |
| Priority: | P2 | Keywords: | FixedInTrunk, PatchAvailable |
| Version: | 2.4.3 | ||
| Target Milestone: | --- | ||
| Hardware: | PC | ||
| OS: | Linux | ||
| Attachments: | An improved log message patch to make the diagnosis easier | ||
Thank you for the report. I agree with your analysis. This has been applied to trunk. http://svn.apache.org/viewvc?view=revision&sortby=date&revision=1433613 If nobody complains, I will propose for backport for 2.4.x in the coming days. Proposed for backport to v2.4. Backported to v2.4.5. |
Created attachment 29828 [details] An improved log message patch to make the diagnosis easier Hi, all, Is it possible to change the following log message to be more accurate and explicit that links to the configuration directive? (see the patch attached) "AH00565: request failed: URI too long (longer than xxxx)" Currently, it's kind of confusing without reading the code. I even do not know what the "URI" refers to when saw the log message: I didn't know whether it's a problem on the server side or client side. Having read the code, I know this's because and only because the client's HTTP request-line exceeds the server's "LimitRequestLine" setting. Compared with "LimitRequestLine", the log message of "LimitRequestFieldsize" is very very clear: "AH00561: Request header exceeds LimitRequestFieldSize" Moreover, the current message is not acurrate because request-line also consists of the HTTP method and protocol version besides URI. I wonder whether we can have the same style of logging, which is implemented in the patch. Notice that the logging is at the place where "This is only going to happen if it exceeds the configured limit for a request-line", so the patch is valid. (You can search "HTTP_REQUEST_URI_TOO_LARGE" in the code to verify it) Thanks, Jackie