[15194] in Athena Bugs
Sun license timeout continues to suck
daemon@ATHENA.MIT.EDU (John Hawkinson)
Thu Jun 12 19:55:44 1997
Date: Thu, 12 Jun 1997 19:55:27 -0400
To: f_l@MIT.EDU
Cc: bugs@MIT.EDU, dkk@MIT.EDU, 3partysw@MIT.EDU
From: John Hawkinson <jhawk@MIT.EDU>
This morning there was an extended outage (apparently there was some problem
with The NOC!'s monitoring such that ASO wasn't notified) of ziggy.
During that time, Sunsoft software was very cumbersome to use, because
Sun's fallback mechanism is pretty broken.
I've attached the two transactions from bugs where I reported this previously.
I had a zephyr chat w/ Dot this morning and she indicated that this might
be a moot point because Sun has reworked their licensing somewhat, but it
was unclear when we would be upgrading because of other issues, and it's also
unclear whether these issues are resolved.
Could someone try to find out what the timetable for upgrades is,
and perhaps whether the new software is vulnerable.
Was this problem ever reported to Sun [back in August of last year]?
If not, how come?
Thanks!!
--jhawk
[14592] daemon@ATHENA.MIT.EDU (John Hawkinson) Athena Bugs 08/24/96 23:12 (34 lines)
Subject: sparc 8.0I: /mit/sunsoft/sun4bin/dbx
Date: Sat, 24 Aug 1996 23:12:11 -0400
To: bugs@MIT.EDU
From: John Hawkinson <jhawk@MIT.EDU>
Sun's license manager client software does not seem to properly
detect some kinds of license manager server failures.
For instance, right now ziggy is wedged (it seems to have a full
listen queue and does not respond to TCP SYNs on port 1701, ostensibly
the license manager port).
Running "attach sunsoft; /mit/sunsoft/sun4bin/dbx" just hangs for
minutes.
truss-ing it shows it hanging on:
getmsg(4, 0xEFFFEAAC, 0xEFFFE9F8, 0xEFFFEA04) (sleeping...)
And tcpdump indicates:
23:04:14.930796 portnoy.mit.edu.40707 > ZIGGY.MIT.EDU.1701: S 807195648:807195648(0) win 8760 <mss 1460> (DF)
23:04:20.650711 portnoy.mit.edu.40707 > ZIGGY.MIT.EDU.1701: S 807195648:807195648(0) win 8760 <mss 1460> (DF)
23:04:32.080611 portnoy.mit.edu.40707 > ZIGGY.MIT.EDU.1701: S 807195648:807195648(0) win 8760 <mss 1460> (DF)
23:04:54.940457 portnoy.mit.edu.40707 > ZIGGY.MIT.EDU.1701: S 807195648:807195648(0) win 8760 <mss 1460> (DF)
23:05:40.660539 portnoy.mit.edu.40707 > ZIGGY.MIT.EDU.1701: S 807195648:807195648(0) win 8760 <mss 1460> (DF)
Sun should be flamed for not having their clients detect this broken
condition in a timely fashion. After a minute or 3 you finally get a
prompt, but the timeout takes too long. It should also indicate the
nature of this failure so users can report problems with the license
server.
--jhawk
--[14592]-- (nref = [14593])
[14593] daemon@ATHENA.MIT.EDU (John Hawkinson) Athena Bugs 08/24/96 23:21 (25 lines)
Subject: Re: sparc 8.0I: /mit/sunsoft/sun4bin/dbx
Date: Sat, 24 Aug 1996 23:21:43 -0400
To: bugs@MIT.EDU
In-Reply-To: "[14592] in Athena Bugs"
From: John Hawkinson <jhawk@MIT.EDU>
> Sun should be flamed for not having their clients detect this broken
> condition in a timely fashion. After a minute or 3 you finally get a
> prompt, but the timeout takes too long. It should also indicate the
> nature of this failure so users can report problems with the license
> server.
To clarify, the timeout is two minutes:
tcpdump: listening on le0
23:17:02.513846 portnoy.mit.edu.40711 > ZIGGY.MIT.EDU.1701: S 906011648:906011648(0) win 8760 <mss 1460> (DF)
23:17:05.370376 portnoy.mit.edu.40711 > ZIGGY.MIT.EDU.1701: S 906011648:906011648(0) win 8760 <mss 1460> (DF)
23:17:11.090481 portnoy.mit.edu.40711 > ZIGGY.MIT.EDU.1701: S 906011648:906011648(0) win 8760 <mss 1460> (DF)
23:17:22.520391 portnoy.mit.edu.40711 > ZIGGY.MIT.EDU.1701: S 906011648:906011648(0) win 8760 <mss 1460> (DF)
23:17:45.380373 portnoy.mit.edu.40711 > ZIGGY.MIT.EDU.1701: S 906011648:906011648(0) win 8760 <mss 1460> (DF)
23:18:31.100391 portnoy.mit.edu.40711 > ZIGGY.MIT.EDU.1701: S 906011648:906011648(0) win 8760 <mss 1460> (DF)
23:19:02.452087 portnoy.mit.edu.40711 > ZIGGY.MIT.EDU.1701: R 906011649:906011649(0) win 8760 (DF)
23:19:02.463526 portnoy.mit.edu.40712 > GOOSHI.MIT.EDU.1701: S 921435648:921435648(0) win 8760 <mss 1460> (DF)
23:19:02.466509 GOOSHI.MIT.EDU.1701 > portnoy.mit.edu.40712: S 673563648:673563648(0) ack 921435649 win 8760 <mss 1460> (DF)
--jhawk
--[14593]-- (pref = [14592])
