Re: (ASCEND) no surprise: p42 leaks

[Bruce Robertson <bruce@greatbasin.net>]

We're seeing what appears to be a very slow leak on our most heavily loaded
Max (first one in the hunt group, and *always* has 47 channels in use, even at
4am).  I haven't reported it to support yet, because I'm waiting to see if
the memory really does go to zero and cause a hang or reboot.  Here's the stats
so far:

Validating free lists...
0 errors; 261 buffers; 442048 bytes memory free

system uptime: up 14 days, 0 hours, 13 minutes, 7 seconds

This Max has OSPF running, and stacking enabled.  Even with a potential memory
leak, we're extremely happy with this revision, as it's the first one ever that
has a usable OSPF implementation.

It does seem to be related to load, because here's the next two Maxen in the
hunt group.  They've all been up for the same 14 days.

Validating free lists...
0 errors; 225 buffers; 594560 bytes memory free

Validating free lists...
0 errors; 377 buffers; 741280 bytes memory free

Meanwhile, here's a much more lightly loaded Max in one of our remote towns -
still using OSPF, but no stacking since it's a single Max.  This Max has only
a single channelized T1 attached.

Validating free lists...
0 errors; 229 buffers; 1530528 bytes memory free

--
Bruce Robertson, President/CEO				     +1-702-348-7299
Great Basin Internet Services, Inc.			fax: +1-702-348-9412
PGP Key fingerprint = 03 2D A8 4A 37 07 FB 53  EB C9 02 5C E5 10 66 F2


++ Ascend Users Mailing List ++
To unsubscribe:	send unsubscribe to ascend-users-request@bungi.com
To get FAQ'd:	<http://www.nealis.net/ascend/faq>

---

References:

• Re: (ASCEND) no surprise: p42 leaks
• From: MMedwid@symantec.com (Michael Medwid)

---

• Prev by Date: (ASCEND) Session Timers
• Next by Date: Re: (ASCEND) Ascend bashing- actually these guys aren't all that bad
• Prev by thread: Re: (ASCEND) no surprise: p42 leaks
• Next by thread: (ASCEND) 128K connection from Pipe50 to Cisco3620 BRI
• Index(es):
  • Main
  • Thread