[Search for users]
[Overall Top Noters]
[List of all Conferences]
[Download this site]
| Title: | DECWINDOWS 26-JAN-89 to 29-NOV-90 |
| Notice: | See 1639.0 for VMS V5.3 kit; 2043.0 for 5.4 IFT kit |
| Moderator: | STAR::VATNE |
|
| Created: | Mon Oct 30 1989 |
| Last Modified: | Mon Dec 31 1990 |
| Last Successful Update: | Fri Jun 06 1997 |
| Number of topics: | 3726 |
| Total number of notes: | 19516 |
1447.0. "Strange resource wait state" by RATH::BROOKE (Justice Ongar Twilight) Fri Sep 15 1989 13:02
[Also posted in FIREFOX notesfile]
Help!
Just taken delivery of a new 3540. I've been autogenning it into a reasonable
state of health, but I get some very strange things happening. Occasionally,
when closing an SDC DECwindows (VMS v5.2) session down, the root window doesn't
get reset, and the 3540 then goes into a strange resource wait state. I ran
VPA and it gave the following cryptic advice:
VPA V2.0 Analysis RATH (VAXstation 3540) PAGE 1
Report Friday 15SEP89 00:00 to 15:54
CONCLUSION 1.
{R0280}
The following image(s) were hung in an Mwait state for
a period of at least two minutes. Attempt to alleviate
the problem by providing more of the resource which was
in short supply, if possible. Generally, processes
should not be delayed for this long of a period, thus
indicating a possible unusual problem for which an
expert may be required.
CONDITIONS
1. IMAGE_HUNG_IN_MWAIT_NOT_RWAST .EQ. 1.00
2. OCCURRENCES .GE. 1
EVIDENCE
Event flag --------Time Range------- No. of
User Image Code First time Last time records
------------ --------- ------------- ------------------------------- -------
BROOKE DB3-40 RSN$_MPWBUSY 15-SEP 15:36:00 15-SEP 15:36:00 1
Event flag description: modified page writer is busy
SYSTEM ********* RSN$_MPWBUSY 15-SEP 15:36:00 15-SEP 15:36:00 1
Event flag description: modified page writer is busy
VPA V2.0 Analysis RATH (VAXstation 3540) PAGE 2
Report Friday 15SEP89 00:00 to 15:54
CONCLUSION 2.
{R0010}
The system fault rate for VMS is over 2 faults per
second for the following time periods. Performance can
be improved for the whole system if the VMS fault rate
can be reduced.
Increase the working set size for VMS (SYSMWCNT) to
reduce the system fault rate. Do this by adding an
entry in MODPARAMS.DAT similar to "ADD_SYSMWCNT = 100",
and running AUTOGEN. The "100" is an initial
guideline.
Total number of samples supporting this conclusion: 27
CONDITIONS
1. SYSTEM_FAULT_RATE .GE. 3.00
2. OCCURRENCES .GE. 10
EVIDENCE
System fault
rate Time of occurrence
------------ ------------------
10.59 15-SEP 12:38:00
4.47 15-SEP 12:40:00
7.46 15-SEP 12:44:00
5.71 15-SEP 12:48:00
4.59 15-SEP 12:58:00
5.59 15-SEP 13:00:00
6.84 15-SEP 13:10:00
7.73 15-SEP 13:14:00
7.13 15-SEP 13:20:00
5.27 15-SEP 13:28:00
8.78 15-SEP 13:30:00
7.31 15-SEP 13:40:00
14.28 15-SEP 13:42:00
16.49 15-SEP 13:44:00
7.18 15-SEP 13:48:00
15.50 15-SEP 13:50:00
4.15 15-SEP 13:56:00
7.22 15-SEP 13:58:00
12.55 15-SEP 14:00:00
3.20 15-SEP 14:02:00
17.06 15-SEP 14:08:00
13.58 15-SEP 14:10:00
3.29 15-SEP 14:12:00
16.44 15-SEP 15:14:28
3.46 15-SEP 15:20:00
13.57 15-SEP 15:24:00
4.49 15-SEP 15:32:00
VPA V2.0 Analysis RATH (VAXstation 3540) PAGE 3
Report Friday 15SEP89 00:00 to 15:54
ANALYSIS SUMMARY for node RATH
Number of Records Processed.......................60
Number of Records satisfying rule conditions......28
Number of Records not satisfying rule conditions..32
Number of Conclusions.............................2
User Command: ADVISE/REPORT=ANAL/NODE=RATH/OUT=XYZ
Anybody seen this one before/know what's happening?
| T.R | Title | User | Personal
Name | Date | Lines |
|---|
| 1447.1 | | PSW::WINALSKI | Careful with that VAX, Eugene | Fri Sep 15 1989 18:21 | 4 |
| I think you need more pagefile space.
--PSW
|
| 1447.2 | | RATH::BROOKE | Justice Ongar Twilight | Wed Sep 20 1989 16:53 | 5 |
| Thanks; increasing pagefile at least got it to the stage where I could
log out and log back in again - always an advantage, especially in a
12mip machine....
|