Bug 649194

Summary:	package.mask >=sys-libs/db-6.1 because of test failure, but only because wrong tcl version?
Product:	Gentoo Linux	Reporter:	Ulenrich <ulenrich>
Component:	Current packages	Assignee:	Gentoo's Team for Core System packages <base-system>
Status:	RESOLVED OBSOLETE
Severity:	normal	CC:	pacho
Priority:	Normal
Version:	unspecified
Hardware:	All
OS:	Linux
Whiteboard:
Package list:		Runtime testing required:	---

Description Ulenrich 2018-03-01 14:22:39 UTC

I am wondering why the package.mask sys-libs/db-6.2*
is so old from 2014. I noticed the Url of oracle says:
http://download.oracle.com/otndocs/products/berkeleydb/html/changelog_6_2.html
The test suite now requires Tcl 8.6 or greater
---
but our ebuild only requests:
DEPEND="tcl? ( >=dev-lang/tcl-8.5.15-r1:0=[${MULTILIB_USEDEP}] )
        test? ( >=dev-lang/tcl-8.5.15-r1:0=[${MULTILIB_USEDEP}] )
---
Also I see the files/db-6.0.20-test-link.patch may be too old ?

Comment 1 Lars Wendler (Polynomial-C) (RETIRED) gentoo-dev

2018-03-01 14:38:54 UTC

AFAIK it's not only because of failing tests but also because Oracle moved the license of >=db-6.1 from Sleepycat to AGPL-3 which makes it incompatible with lots of other software.
Perhaps we should reflect that in the mask reason as well.

Comment 2 Ulenrich 2018-03-01 17:48:37 UTC

@Lars, I would rather like to have an additional  
USE agpl_accept
as a license warning
than an obscured package.mask for 
>sys-libs/db-6.1

Comment 3 Robin Johnson archtester

2018-03-01 18:45:06 UTC

No, it's ONLY masked for failing tests. It is not masked because of AGPL.
I will update the Tcl dep, but it does still fail here with Tcl-8.6.5.

Comment 4 Ulenrich 2018-03-02 20:45:52 UTC

@Robin thanx for explanation. 
And I just reported, because both: it is an old mask and this test dependency did not fit. Feel free to close the bug then.

Comment 5 Robin Johnson archtester

2018-03-02 21:35:03 UTC

I re-ran the tests w/ a newer system, and got only 2 failures now:
UNEXPECTED OUTPUT: FAIL: condition {[stat_field $env rep_stat "Startup complete"]} not achieved in 20 seconds.
UNEXPECTED OUTPUT: BDB0068 DB_LOCK_DEADLOCK: Locker killed to resolve a deadlock

took ~3 hours on an unloaded i7 X980 system on SSD.

Running it again for good measure.

Comment 6 Robin Johnson archtester

2018-03-05 19:33:29 UTC

Next run (again, -j12, 3 hour duration:
UNEXPECTED OUTPUT: db put:permission denied

So i'm wondering if the real bug is around concurrency.

Running again @ -j1:
UNEXPECTED OUTPUT: db put:permission denied

Comment 7 Sam James archtester

2023-01-09 13:24:10 UTC

We last-rited and removed this version.