[2161] in java-interest
putmsg multiply defined in java VM (JDK) (fwd)
daemon@ATHENA.MIT.EDU (Steve V. Yalovitser)
Tue Sep 26 12:43:35 1995
From: "Steve V. Yalovitser" <root@cyberpunk.com>
To: java-interest@java.Eng.Sun.COM
Date: Tue, 26 Sep 1995 10:25:47 +0900 (EDT)
Hiya,
Ive recently run into a lot of core dumps from failed asserts around
iomgr.c(line 628) the call stack's last call is putmsg before panic. The
problem is
that putmsg is in libc.a as well. The cores result because I think the
_wrong_putmsg is being called. In particular one core has a stack which has:
panic()
putmsg()
t_sndudata()
clnt_dg_call()
_nis_finddirectory_r()
The putmsg above is probably called out from t_sndudata (a system call). The
above is repeatable on a few different architectures running solaris 2.4. It
isrepeatable easily by doing any calls that require hostname lookup (for
instance through InetAddress.java). Thanks in advance for any help.
-
Note to Sun employees: this is an EXTERNAL mailing list!
Info: send 'help' to java-interest-request@java.sun.com
