"GOT", but the "O" is a cute, smiling pufferfish. Index | Thread | Search

From:
Stefan Sperling <stsp@stsp.name>
Subject:
Re: gotd crash while requesting non-existent commit id
To:
Mikhail <mp39590@gmail.com>
Cc:
gameoftrees@openbsd.org
Date:
Sun, 4 Dec 2022 13:44:38 +0100

Download raw body.

Thread
On Sun, Dec 04, 2022 at 03:15:58PM +0300, Mikhail wrote:
> On Fri, Nov 04, 2022 at 01:47:43PM +0100, Stefan Sperling wrote:
> > On Fri, Nov 04, 2022 at 03:37:12PM +0300, Mikhail wrote:
> > > While doing some kind of "manual stress testing" with gotd it crashed
> > > for me. Crashes are repeatable.
> > 
> > Thanks, there was a simple coding error.
> > Should be fixed by commit 36c7cfbb2a9b646bfb1658fca4e34bc63a46ec42
> 
> Here is a test case for this bug.
> 
> Can it be useful?

Yes, more test cases are always good to have, thanks! A question below:

> --- /dev/null
> +++ regress/gotd/repo_req_wrong_commit.sh

> +	# Before 36c7cfbb2a gotd was crashing after invalid commit in "want"
> +	# line, here we check that we can successfully stop working gotd.
> +	$GOTD_STOP_CMD
> +	ret=$?

GOTD_STOP_CMD requires root privileges to succeed, but gotd test scripts
run as GOTD_TEST_USER. So this command should always fail. If it does not
fail then there is a problem.

I am wondering if this check could be done in the Makefile instead, where
GOTD_STOP_CMD runs as root?