https://bz.apache.org/bugzilla/show_bug.cgi?id=70080
Bug ID: 70080
Summary: apachectl configtest does not warn about deprecated
variables
Product: Apache httpd-2
Version: 2.4.67
Hardware: PC
OS: Linux
Status: NEW
Severity: normal
Priority: P2
Component: Runtime Config
Assignee: [email protected]
Reporter: [email protected]
Target Milestone: ---
I have been switching from mpm_prefork to mpm_event.
Now, I had "MaxRequestsPerChild 1" in my conf-enabled that I missed.
What this lead to was that if I created a simple static file: `dd
if=/dev/urandom of=/var/www/html/test-200k.bin bs=200K count=1`
And then repeatedly downloaded it: `while curl --verbose --fail
http://1.2.3.4/test-200k.bin ; do echo "a"; done`
I'd often get this:
* transfer closed with outstanding read data remaining
* Closing connection 0
curl: (18) transfer closed with outstanding read data remaining
Despite Apache not logging any error.
Setting `MaxConnectionsPerChild 0` fixed it and I could not understand why
especially since that seemed to already be set by default - what was the issue
was the aforementioned earlier "MaxRequestsPerChild 1" definition
Now whether any of the above is a bug or not, you tell me, and I will open a
new report if needed, but what this report is actually about is the fact that
`apachectl configtest` does not warn about such deprecations.
In mpm_common.c, I see this block:
```
if (!strcasecmp(cmd->cmd->name, "MaxRequestsPerChild")) {
ap_log_error(APLOG_MARK, APLOG_INFO, 0, NULL, APLOGNO(00545)
"MaxRequestsPerChild is deprecated, use "
"MaxConnectionsPerChild instead.");
}
```
But I have not once ever seen this warning - under which conditions was I
suppose to see it?
Doing a config test happily reports that the syntax is correct:
# apachectl configtest
Syntax OK
I believe configtest should report deprecations, if any exist.
I am not aware of any other mechanism which would warn against them.
--
You are receiving this mail because:
You are the assignee for the bug.
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
