Error binding port, ICL initialization failed

[David Wei]

Clovertechers:

First time to use inter-site routing on CIS 5.8.4, everything looks ok. however, when re-start the receiving process, we got the following:

[prod:prod:INFO/0: star_ice_out:02/21/2012 20:05:39] Starting protocol thread star_ice_out as tid 3.

[icl :tcpi:ERR /0: star_ice_out:02/21/2012 20:05:39] Error binding port 12602

[pd  :thrd:ERR /0: star_ice_out:02/21/2012 20:05:39] ICL initialization failed

PANIC: “0”

PANIC: Calling “pti” for thread star_adt_cmd

---

Scheduler State

---

Thread Events     State      Priority Runnable  PT Msgs

  0      0   SCHED_IDLE         0       0       0,0,0

….

---

Outstanding Pthread User Data Msgs

---

pmq: 0x0xa6ada10

Count   : 0

Head    : 0x(nil)

Tail    : 0x(nil)

PANIC: Calling “dbi shutdown” for thread star_adt_cmd

PANIC: Calling “dbi shutdown” for thread star_adt_xlate

PANIC: Calling “clean up ODBC” for thread star_adt_cmd

PANIC: Calling “dbi shutdown” for thread star_adt_loopout_s4

PANIC: Calling “dbi shutdown” for thread star_ice_out

PANIC: Process panic—engine going down

PANIC: assertion ‘0’ failed at protocolThread.cpp/521

Tried to bring up the thread / process and a couple of minutes later, it came up.

Seems this happenes everytime we bounce the process. eventually it’ll come up w/o any other action.

Does anyone know why it’s like that?

thanks,

David

York Central Hospital

Topic

[Bob Richardson]

Greetings,

It pays to read ALL the release notes for any Integrator Release version!

You will find in the 5.8.4.0 Release notes under Known Issues note # 6671 titled “Engine panics when inter-site routing port is re-binded“.

It seems to suggest that you have to wait for “a short time” (whatever that is)

before starting up the public receiver thread.

We are currently evaluating the Revision 5 patches now and inter-site routing is on our radar to check out.  This problem is still listed under the Known Issues heading – same item number.

You might want to consider using the old tried but true solution known as “local hosting”.  Of course we have an unlimited thread count license and maybe that is your concern here?  not consuming your thread counts?

If not then the “local hosting” method should work for you.

Let this forum know if you need help on how to do that configuration (local hosting).

Good luck, good hunting!

[David Wei]

Thanks Bob,

We used to use localhost to loopin/out and there’s no issue at all.

Just want to try something new.

Next time, I’ll read the release notes first.

Take care,

David

[Bob Richardson]

David,

If you get the inter-site routing to work and remain stable let this forum know.

I would still like to use it if inter-site routing can be guaranteed to work as advertised.  And guarantees reliable delivery of messages especially in the health care setting where patient safety is concerned.

Thanks.

BobR

[David Barr]

You can use the netstat command to figure out when the thread is safe to restart without a panic. Just run netstat -an and look for your port number. Wait until the port number disappears or is no longer in a CLOSE_WAIT state.

[Author]

Replies