[omniORB] [omniNotify] coring notifd?

Robert E. Gruber gruber@research.att.com
Wed, 17 Oct 2001 11:17:29 -0400


Please have the person upgrade to omniNotify version 1.1,
which is available at
http://www.research.att.com/~ready/omniNotify/nfy_download.html
This will solve the problem.

-- Bob


-----Original Message-----
From: owner-omniorb-list@uk.research.att.com
[mailto:owner-omniorb-list@uk.research.att.com]On Behalf Of Rakshit
Simha
Sent: Wednesday, October 17, 2001 11:01 AM
To: omniorb-list@uk.research.att.com
Subject: [omniORB] [omniNotify] coring notifd?


Hi folks,

I'm posting this on behalf of a colleague whose emails (for some reason)
are not making it into the mailing list.  Please direct responses to
the mailing list and/or ehung@orchestream.com.

---begin---
Hi,

We have been experiencing the following situation (omniORB 3.0.2,
omniNotify 1.0, Solaris 8, 32 bit sparc, Sun CC Forte 6U1)

It seems that a thread is created for each batch of events (with the
thread id being incremented for each thread). When the thread id
reaches 1024+ (RDI_TH_ARRAY_SZ defined in RDIChannelUtil.h), notifd
cores when an assertion fails on the thread id (line 84
RDIChannelUtil.h).

We have a push supplier and push consumer connected to notifd. The
supplier for each batch sets up a connection, pushes the batch and
closes the connection. The omniConnectionBroken exception appears to
be thrown for each thread.

Has anyone else experienced such a situation?

Elaine
---end---

Regards,
Rak.
---
Rakshit Simha                       Phone: (613) 599-2300 x8958
Senior Software Designer            Fax  : (613) 599-2330
Orchestream Americas Corp.          Email: rsimha@orchestream.com
Kanata, Canada                      http://www.orchestream.com/