- This topic has 4 replies, 3 voices, and was last updated 7 years, 6 months ago by
.
-
Topic
-
I realize that the information provided by a PANIC is next to useless but I have seen this type of panic 3 times in the last 3 weeks. With each time the immediate solution was to restart the process. Server: Windows Server 2008
Cloverleaf: 5.8.5
Here are the first two lines from the panic:
PANIC: “((ev)->ev_flags & (0x00000001)) || ((ev)->ev_flags & (0x00000002)) || ((ev)->ev_flags & (0x00000004))”
PANIC: Calling “pti” for thread p_adt_cmd
Everything after here is the typical information that is of no value to the customer.
Has anyone seen this before?
David Coffey
-
Replies
-
-
Hi David,
I can’t make anything from these panic-dumps either, even the first 2 lines, although I haven’t seen panics for a loooong time (Linux…).
So standard questions:
* How many threads inside the process that paniced?
* Enough memory?
* Large messages or large quantity of messages?
Many moons ago (CL 3.7/5.2 on HP-UX) we would see panics when we had too many threads in one process. Now we sometimes see it when we configure something totally wrong.
Zuyderland Medisch Centrum; Heerlen/Sittard; The Netherlands
-
I prefer to develop on Unix/Linux myself but had no input into the OS for Cloverleaf. Long ago, far, far away in another galaxy I worked at a shop that ran CAI TDM (later know as Sybase e_biz) on AIX and we went 500+ days w/o a reboot or restart on the product.
As to the thread count? 45. I have not looked this up recently but I do not think I am pushing the maximum recommended count for a process. Does anyone know different?
This process handles primarily ADT with a few threads that merge up outbound results and/or orders to some clinics. I do not know the answer to your question about message volumes or message size. Worth looking into.
David
-
-
-
- The forum ‘Cloverleaf’ is closed to new topics and replies.
