Re: krb5-1.22

On Mon, 2025-08-11 at 20:07 -0500, Bruce Dubbs wrote:
> Currently we are looking for the krb5 package at
> https://kerberos.org but that site 
> seems to be down.  With a little searching I found
> https://web.mit.edu/kerberos/dist/
> and was able to download from
> 
> https://web.mit.edu/kerberos/dist/krb5/1.22/krb5-1.22.tar.gz
> 
> Looking at https://web.mit.edu/kerberos/krb5-1.22/krb5-1.22.html, it
> still says to 
> get the package at https://kerberos.org/dist/index.html
> 
> We can wait a day or two to see if kerberos.org comes back up or I can
> change the 
> package to point to web.mit.edu/kerberos.
> 
> The current download problem may be related in some way to the fact
> that
> https://ftp.gnu.org/gnu/ has also been down since last night.  That
> affects a lot of 
> packages in LFS as well as 26 packages in BLFS.
> 
> Amy thoughts on how to address this?
> 
> -----
> 
> Also I built krb5-1.22 without the patch or the sed.  I found two instances of 
> 'failed' in the tests:
> 
> _kerberos._udp.athena.mit.edu. 1800 IN SRV   0 0 88 kerberos-1.mit.edu.
> _kerberos._udp.athena.mit.edu. 1800 IN SRV   0 0 88 kerberos-2.mit.edu.
> _kerberos._udp.athena.mit.edu. 1800 IN SRV   0 0 88 kerberos.mit.edu.
> in module_locate_server
> ran off end of plugin list
> looking in krb5.conf for realm ATHENA.MIT.EDU entry kdc; ports 88,0
> config file lookup failed: Profile relation not found
> 3 servers:
> 0: h:kerberos-1.mit.edu. t:udp p:88 m:-1 P:
> 1: h:kerberos-2.mit.edu. t:udp p:88 m:-1 P:
> 2: h:kerberos.mit.edu. t:udp p:88 m:-1 P:
> 
> The best I can tell, this is related to the outage of kerberos.org.
> 
> The second failure instance has output:
> 
> Test 3: btree: small key, big data pairs
>      page size 512
>      page size 16384
>      page size 65536
> Test 3: recno: big data pairs
>      page size 512
>      page size 16384
>      page size 65536
> Test 4: recno: random entries
> Test 5: recno: reverse order entries
> Test 6: recno: alternating order entries
> Test 7: btree, recno: delete cursor record
> Test 8: btree: repeated small key, big data pairs
>     1 of 10
>     2 of 10
>     3 of 10
>     4 of 10
>     5 of 10
>     6 of 10
>     7 of 10
>     8 of 10 (comparison)
>     9 of 10
>     10 of 10
> Test 9: btree: duplicate keys
> Test 10: btree, recno: test cursor flag use
> Test 11: recno: reverse order insert
> Test 12: btree: lots of keys, small page size
> test12: type btree: failed
> 
> How should this be handled?  I'm not sure if this is related to the
> comment we have 
> not that says: "One test, t_kadm5srv, is known to fail."
> 
> How should this be handled?

This one should be handled by the sed.  See
https://krbdev.mit.edu/rt/Ticket/Display.html?id=9129 (the browser will
report "the certificate is too old" but you can ignore the warning).

The sed in the book removes this test.


-- 
Xi Ruoyao <[email protected]>

-- 
http://lists.linuxfromscratch.org/sympa/info/blfs-dev
Unsubscribe: See the above information page