Menu
▾
▴
bproc-users — General discussion about BProc.
You can subscribe to this list here.
| 2001 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
(25) |
Nov
|
Dec
(22) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2002 |
Jan
(13) |
Feb
(22) |
Mar
(39) |
Apr
(10) |
May
(26) |
Jun
(23) |
Jul
(38) |
Aug
(20) |
Sep
(27) |
Oct
(76) |
Nov
(32) |
Dec
(11) |
| 2003 |
Jan
(8) |
Feb
(23) |
Mar
(12) |
Apr
(39) |
May
(1) |
Jun
(48) |
Jul
(35) |
Aug
(15) |
Sep
(60) |
Oct
(27) |
Nov
(9) |
Dec
(32) |
| 2004 |
Jan
(8) |
Feb
(16) |
Mar
(40) |
Apr
(25) |
May
(12) |
Jun
(33) |
Jul
(49) |
Aug
(39) |
Sep
(26) |
Oct
(47) |
Nov
(26) |
Dec
(36) |
| 2005 |
Jan
(29) |
Feb
(15) |
Mar
(22) |
Apr
(1) |
May
(8) |
Jun
(32) |
Jul
(11) |
Aug
(17) |
Sep
(9) |
Oct
(7) |
Nov
(15) |
Dec
|
|
From: Jack N. <jj...@pa...> - 2002-09-17 20:10:27
|
On Tue, Sep 17, 2002 at 03:28:36PM -0400, er...@he... wrote: > On Tue, Sep 17, 2002 at 11:06:24AM -0400, Jack Neely wrote: > > Folks, > > > > I've gotten a but further but I've hit some other things I need to > > figure out. > > > > Nodes are booting and getting into phase 2 right where bpslave is > > executed. (bpslave -d -i -c /rootfs -p 10001 192.168.1.10) Which > > returns with "Error getting file cache FD: Invalid argument". So, I'm > > thinking that the node is having problems loading libraries or kernel > > modules or the like. > > I sounds like you got BProc built w/o file cache support built in. > Make sure FILEREQ:=y is in Makefile.conf. (See the release notes.) > If you're still not getting it, then it sounds like I screwed up > somewehere. > > > I haven't patched glibc yet. (Hoping to avoid that but...) Also, I > > don't seem to have the patch for modutils mentioned in the docs on the > > BProc web site. Is one of these the source of my problem? > > You will need that. You can grab a patched (but somewhat old) library > off www.clustermatic.org if you don't feel like rebuilding it. > > > FYI, bproc_common.h isn't getting installed in the bproc.spec file. > > Needs to be added to the %files section. > > Thanks. > > - Erik Well, I got a new glibc built with the bproc patches but still am not getting any further. I do have things compiled with FILEREQ:=y in the Makefile.conf. I've also checked to make sure I've got all the stuff compiled into my kernel that's on the docs on sourceforge. I've not tried a patched modutils package yet. I've looked at the older package at Clustermatic. I'll have to port it forward but doesn't look too difficult. If there's any info I can provide please ask. Jack Neely -- Jack Neely <sl...@qu...> Linux Realm Kit Administration and Development PAMS Computer Operations at NC State University GPG Fingerprint: 1917 5AC1 E828 9337 7AA4 EA6B 213B 765F 3B6A 5B89 |
|
From: <er...@he...> - 2002-09-17 19:40:03
|
On Tue, Sep 17, 2002 at 11:06:24AM -0400, Jack Neely wrote: > Folks, > > I've gotten a but further but I've hit some other things I need to > figure out. > > Nodes are booting and getting into phase 2 right where bpslave is > executed. (bpslave -d -i -c /rootfs -p 10001 192.168.1.10) Which > returns with "Error getting file cache FD: Invalid argument". So, I'm > thinking that the node is having problems loading libraries or kernel > modules or the like. I sounds like you got BProc built w/o file cache support built in. Make sure FILEREQ:=y is in Makefile.conf. (See the release notes.) If you're still not getting it, then it sounds like I screwed up somewehere. > I haven't patched glibc yet. (Hoping to avoid that but...) Also, I > don't seem to have the patch for modutils mentioned in the docs on the > BProc web site. Is one of these the source of my problem? You will need that. You can grab a patched (but somewhat old) library off www.clustermatic.org if you don't feel like rebuilding it. > FYI, bproc_common.h isn't getting installed in the bproc.spec file. > Needs to be added to the %files section. Thanks. - Erik |
|
From: Nicholas H. <he...@ce...> - 2002-09-17 18:55:37
|
Here is the last bit of an strace of the process:
stat64("/usr/java/jdk1.3.1_03/jre/classes", 0xbfffbe24) = -1 ENOENT (No
such file or directory)
mmap2(NULL, 33554432, PROT_READ|PROT_WRITE|PROT_EXEC,
MAP_PRIVATE|MAP_ANONYMOUS|MAP_NORESERVE, -1, 0) = 0x415bb000
mmap2(0x415bb000, 163840, PROT_READ|PROT_WRITE|PROT_EXEC,
MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0x415bb000
mmap2(NULL, 524288, PROT_READ|PROT_WRITE|PROT_EXEC,
MAP_PRIVATE|MAP_ANONYMOUS|MAP_NORESERVE, -1, 0) = 0x435bb000
mmap2(0x435bb000, 4096, PROT_READ|PROT_WRITE|PROT_EXEC,
MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0x435bb000
brk(0x8057000) = 0x8057000
mmap2(NULL, 163840, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x4363b000
mmap2(NULL, 100663296, PROT_READ|PROT_WRITE|PROT_EXEC,
MAP_PRIVATE|MAP_ANONYMOUS|MAP_NORESERVE, -1, 0) = 0x43663000
munmap(0x43663000, 100663296) = 0
mmap2(NULL, 100728832, PROT_READ|PROT_WRITE|PROT_EXEC,
MAP_PRIVATE|MAP_ANONYMOUS|MAP_NORESERVE, -1, 0) = 0x43663000
munmap(0x43663000, 53248) = 0
munmap(0x49670000, 12288) = 0
mmap2(0x43670000, 655360, PROT_READ|PROT_WRITE|PROT_EXEC,
MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0x43670000
mmap2(0x47670000, 1048576, PROT_READ|PROT_WRITE|PROT_EXEC,
MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0x47670000
mmap2(NULL, 65536, PROT_READ|PROT_WRITE|PROT_EXEC,
MAP_PRIVATE|MAP_ANONYMOUS|MAP_NORESERVE, -1, 0) = 0x49670000
mmap2(0x49670000, 4096, PROT_READ|PROT_WRITE|PROT_EXEC,
MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0x49670000
mmap2(0x438f0000, 1441792, PROT_READ|PROT_WRITE|PROT_EXEC,
MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0x438f0000
mmap2(NULL, 126976, PROT_READ|PROT_WRITE|PROT_EXEC,
MAP_PRIVATE|MAP_ANONYMOUS|MAP_NORESERVE, -1, 0) = 0x49680000
mmap2(0x49680000, 4096, PROT_READ|PROT_WRITE|PROT_EXEC,
MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0x49680000
brk(0x8061000) = 0x8061000
mmap2(NULL, 196608, PROT_READ|PROT_WRITE|PROT_EXEC,
MAP_PRIVATE|MAP_ANONYMOUS|MAP_NORESERVE, -1, 0) = 0x4969f000
mmap2(0x4969f000, 8192, PROT_READ|PROT_WRITE|PROT_EXEC,
MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0x4969f000
mmap2(0x496bf000, 4096, PROT_READ|PROT_WRITE|PROT_EXEC,
MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0x496bf000
brk(0x806f000) = 0x806f000
uname({sys="Linux", node="node9", ...}) = 0
gettimeofday({1032288738, 518242}, NULL) = 0
brk(0x8078000) = 0x8078000
brk(0x8081000) = 0x8081000
gettimeofday({1032288738, 538011}, NULL) = 0
brk(0x8084000) = 0x8084000
open(".hotspot_compiler", O_RDONLY) = -1 ENOENT (No such file or
directory)
mmap2(NULL, 454656, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x496cf000
brk(0x808d000) = 0x808d000
munmap(0x496cf000, 454656) = 0
mmap2(NULL, 454656, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x496cf000
munmap(0x496cf000, 454656) = 0
mmap2(NULL, 454656, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x496cf000
munmap(0x496cf000, 454656) = 0
mmap2(NULL, 454656, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x496cf000
munmap(0x496cf000, 454656) = 0
mmap2(NULL, 454656, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x496cf000
munmap(0x496cf000, 454656) = 0
mmap2(NULL, 454656, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x496cf000
munmap(0x496cf000, 454656) = 0
mmap2(NULL, 454656, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x496cf000
munmap(0x496cf000, 454656) = 0
mmap2(NULL, 454656, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x496cf000
munmap(0x496cf000, 454656) = 0
mmap2(NULL, 454656, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x496cf000
munmap(0x496cf000, 454656) = 0
mmap2(NULL, 454656, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x496cf000
munmap(0x496cf000, 454656) = 0
mmap2(NULL, 454656, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x496cf000
munmap(0x496cf000, 454656) = 0
mmap2(NULL, 454656, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x496cf000
munmap(0x496cf000, 454656) = 0
mmap2(NULL, 454656, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x496cf000
munmap(0x496cf000, 454656) = 0
mmap2(NULL, 454656, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x496cf000
munmap(0x496cf000, 454656) = 0
mmap2(NULL, 454656, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x496cf000
munmap(0x496cf000, 454656) = 0
mmap2(NULL, 454656, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x496cf000
munmap(0x496cf000, 454656) = 0
mmap2(NULL, 454656, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x496cf000
munmap(0x496cf000, 454656) = 0
mmap2(NULL, 454656, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x496cf000
munmap(0x496cf000, 454656) = 0
mmap2(NULL, 454656, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x496cf000
munmap(0x496cf000, 454656) = 0
mmap2(NULL, 454656, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x496cf000
munmap(0x496cf000, 454656) = 0
mmap2(NULL, 454656, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x496cf000
munmap(0x496cf000, 454656) = 0
mmap2(NULL, 454656, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x496cf000
munmap(0x496cf000, 454656) = 0
mmap2(NULL, 454656, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x496cf000
munmap(0x496cf000, 454656) = 0
mmap2(NULL, 454656, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x496cf000
munmap(0x496cf000, 454656) = 0
mmap2(NULL, 454656, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x496cf000
munmap(0x496cf000, 454656) = 0
mmap2(NULL, 454656, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x496cf000
munmap(0x496cf000, 454656) = 0
mmap2(NULL, 454656, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x496cf000
munmap(0x496cf000, 454656) = 0
mmap2(NULL, 454656, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x496cf000
munmap(0x496cf000, 454656) = 0
mmap2(NULL, 454656, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x496cf000
munmap(0x496cf000, 454656) = 0
mmap2(NULL, 454656, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x496cf000
munmap(0x496cf000, 454656) = 0
mmap2(NULL, 454656, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x496cf000
munmap(0x496cf000, 454656) = 0
mmap2(NULL, 454656, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x496cf000
munmap(0x496cf000, 454656) = 0
mmap2(NULL, 454656, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x496cf000
munmap(0x496cf000, 454656) = 0
mmap2(NULL, 454656, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x496cf000
munmap(0x496cf000, 454656) = 0
mmap2(NULL, 454656, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x496cf000
munmap(0x496cf000, 454656) = 0
mmap2(NULL, 454656, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x496cf000
munmap(0x496cf000, 454656) = 0
mmap2(NULL, 454656, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x496cf000
munmap(0x496cf000, 454656) = 0
getrlimit(0x3, 0xbfffd5e8) = 0
brk(0x808f000) = 0x808f000
pipe([3, 4]) = 0
clone(child_stack=0x808d690,
flags=CLONE_VM|CLONE_FS|CLONE_FILES|CLONE_SIGHAND) = 17624
write(4, "\241\224\0@\5\0\0\0\214\361Q@\320k2@\200\326\377\277\261"...,
148) = 148
rt_sigprocmask(SIG_SETMASK, NULL, [32], 8) = 0
write(4, "\200\357\2@\0\0\0\0\230\327\377\277\2104\30@\320\254\10"...,
148) = 148
rt_sigprocmask(SIG_SETMASK, NULL, [32], 8) = 0
rt_sigsuspend([] <unfinished ...>
--- SIGRT_0 (Real-time signal 0) ---
<... rt_sigsuspend resumed> ) = -1 EINTR (Interrupted system
call)
sigreturn() = ? (mask now [])
kill(17625, SIGRT_0) = 0
kill(17625, SIGRT_0#) = 0
--- SIGRT_1 (Real-time signal 1) ---
# HotSpot Virtual Machine Error, Internal Error
# Please report this error at
# http://java.sun.com/cgi-bin/bugreport.cgi
#
# Error ID: 4F533F4C494E55580E43505002B0
#
# Problematic Thread: prio=1073837936 tid=0x808acd0 nid=0x44d9 runnable
#
+++ killed by SIGRT_1 +++
|
|
From: Nicholas H. <he...@se...> - 2002-09-17 18:51:08
|
OK-- I understand that bpsh $node java will not work as bproc does not dump multiple threads. I have attempted to make a bproc wrapper that would still allow the user to use bproc and java. The wrapper code attached bproc_moves to the node, and then exec's the java binary supplied. When invoked with the classic VM, this approach works wonderfully. When you try to use the server VM ( all of this using ibm-jdk-1.3.1_03 ), I get the following: [henken@alpha henken]$ ./jw 9 /usr/java/jdk1.3.1_03/bin/java -client -version #[henken@alpha henken]$ # HotSpot Virtual Machine Error, Internal Error # Please report this error at # http://java.sun.com/cgi-bin/bugreport.cgi # # Error ID: 4F533F4C494E55580E43505002B0 # # Problematic Thread: prio=1073837936 tid=0x808acd0 nid=0x44a6 runnable # Any ideas on why this would be happening? Nic -- Nicholas Henke Linux cluster system programmer University of Pennsylvania he...@se... - 215.573.8149 |
|
From: Wilton W. <ww...@ha...> - 2002-09-17 16:11:37
|
You most definatly are running into a problem where the loader from glibc is not patched, (it's not nescessary to patch the whole of glibc only patching the loader is sufficient). Also patching modutils is required if you want to load modules ;) The company I am working for Hard Data Ltd. http://www.harddata.com is aboput to relase updated bproc packages and support for Red Hat 7.3 (i686/athlon) and updated packages for Red Hat 7.2 (alpha/alphaev6/alphaev67) within the next couple of days.. actually as soon as I find some web space and places to put a bug tracking/support pages and the like. We haven't totally agreed on using sourceforge or not. - Wilton On Tue, 17 Sep 2002, Jack Neely wrote: > Folks, > > I've gotten a but further but I've hit some other things I need to > figure out. > > Nodes are booting and getting into phase 2 right where bpslave is > executed. (bpslave -d -i -c /rootfs -p 10001 192.168.1.10) Which > returns with "Error getting file cache FD: Invalid argument". So, I'm > thinking that the node is having problems loading libraries or kernel > modules or the like. > > I haven't patched glibc yet. (Hoping to avoid that but...) Also, I > don't seem to have the patch for modutils mentioned in the docs on the > BProc web site. Is one of these the source of my problem? > > FYI, bproc_common.h isn't getting installed in the bproc.spec file. > Needs to be added to the %files section. > > Thanks! > Jack Neely > > -- > Jack Neely <sl...@qu...> > Linux Realm Kit Administration and Development > PAMS Computer Operations at NC State University > GPG Fingerprint: 1917 5AC1 E828 9337 7AA4 EA6B 213B 765F 3B6A 5B89 > > > ------------------------------------------------------- > Sponsored by: AMD - Your access to the experts on Hammer Technology! > Open Source & Linux Developers, register now for the AMD Developer > Symposium. Code: EX8664 http://www.developwithamd.com/developerlab > _______________________________________________ > BProc-users mailing list > BPr...@li... > https://lists.sourceforge.net/lists/listinfo/bproc-users ----[ Wilton William Wong ]--------------------------------------------- 11060-166 Avenue Ph : 01-780-456-9771 High Performance UNIX Edmonton, Alberta FAX: 01-780-456-9772 and Linux Solutions T5X 1Y3, Canada URL: http://www.harddata.com -------------------------------------------------------[ Hard Data Ltd. ]---- |
|
From: Jack N. <jj...@pa...> - 2002-09-17 15:06:32
|
Folks, I've gotten a but further but I've hit some other things I need to figure out. Nodes are booting and getting into phase 2 right where bpslave is executed. (bpslave -d -i -c /rootfs -p 10001 192.168.1.10) Which returns with "Error getting file cache FD: Invalid argument". So, I'm thinking that the node is having problems loading libraries or kernel modules or the like. I haven't patched glibc yet. (Hoping to avoid that but...) Also, I don't seem to have the patch for modutils mentioned in the docs on the BProc web site. Is one of these the source of my problem? FYI, bproc_common.h isn't getting installed in the bproc.spec file. Needs to be added to the %files section. Thanks! Jack Neely -- Jack Neely <sl...@qu...> Linux Realm Kit Administration and Development PAMS Computer Operations at NC State University GPG Fingerprint: 1917 5AC1 E828 9337 7AA4 EA6B 213B 765F 3B6A 5B89 |
|
From: Jack N. <jj...@pa...> - 2002-09-17 00:54:04
|
Wilton, That worked. Thanks a lot for you help. Jack On Mon, Sep 16, 2002 at 05:40:10PM -0600, Wilton Wong wrote: > Change the order: > > -lmodutilobj -lmodutilutil -lmodutils > > has to be > > -lmodutils -lmodutilobj -lmodutilutil > > - Wilton > > On Mon, 16 Sep 2002, Jack Neely wrote: > > > Well...I tired that but no joy. I'm getting pretty much the same error > > messages from gcc. (See attachment.) > > > > Jack > > > > > > On Mon, Sep 16, 2002 at 02:30:57PM -0600, Wilton Wong wrote: > > > modutils-devel has changed in RH 7.3, in the create_init function add > > > "-lmodutilobj -lmodutilutil" to the LIBS environment. > > > > > > beonss is pretty dumb in what it does if you look at it.. there isn't much to > > > update ;) > > > > > > - Wilton > > > > > > > ----[ Wilton William Wong ]--------------------------------------------- > 11060-166 Avenue Ph : 01-780-456-9771 High Performance UNIX > Edmonton, Alberta FAX: 01-780-456-9772 and Linux Solutions > T5X 1Y3, Canada URL: http://www.harddata.com > -------------------------------------------------------[ Hard Data Ltd. ]---- -- Jack Neely <sl...@qu...> Linux Realm Kit Administration and Development PAMS Computer Operations at NC State University GPG Fingerprint: 1917 5AC1 E828 9337 7AA4 EA6B 213B 765F 3B6A 5B89 |
|
From: Wilton W. <ww...@ha...> - 2002-09-16 23:40:26
|
Change the order: -lmodutilobj -lmodutilutil -lmodutils has to be -lmodutils -lmodutilobj -lmodutilutil - Wilton On Mon, 16 Sep 2002, Jack Neely wrote: > Well...I tired that but no joy. I'm getting pretty much the same error > messages from gcc. (See attachment.) > > Jack > > > On Mon, Sep 16, 2002 at 02:30:57PM -0600, Wilton Wong wrote: > > modutils-devel has changed in RH 7.3, in the create_init function add > > "-lmodutilobj -lmodutilutil" to the LIBS environment. > > > > beonss is pretty dumb in what it does if you look at it.. there isn't much to > > update ;) > > > > - Wilton > > > ----[ Wilton William Wong ]--------------------------------------------- 11060-166 Avenue Ph : 01-780-456-9771 High Performance UNIX Edmonton, Alberta FAX: 01-780-456-9772 and Linux Solutions T5X 1Y3, Canada URL: http://www.harddata.com -------------------------------------------------------[ Hard Data Ltd. ]---- |
|
From: Jack N. <jj...@pa...> - 2002-09-16 22:10:08
|
Well...I tired that but no joy. I'm getting pretty much the same error messages from gcc. (See attachment.) Jack On Mon, Sep 16, 2002 at 02:30:57PM -0600, Wilton Wong wrote: > modutils-devel has changed in RH 7.3, in the create_init function add > "-lmodutilobj -lmodutilutil" to the LIBS environment. > > beonss is pretty dumb in what it does if you look at it.. there isn't much to > update ;) > > - Wilton > -- Jack Neely <sl...@qu...> Linux Realm Kit Administration and Development PAMS Computer Operations at NC State University GPG Fingerprint: 1917 5AC1 E828 9337 7AA4 EA6B 213B 765F 3B6A 5B89 |
|
From: Wilton W. <ww...@ha...> - 2002-09-16 20:31:40
|
modutils-devel has changed in RH 7.3, in the create_init function add "-lmodutilobj -lmodutilutil" to the LIBS environment. beonss is pretty dumb in what it does if you look at it.. there isn't much to update ;) - Wilton On Mon, 16 Sep 2002, Jack Neely wrote: > Hi, > > I'm trying to create a Beowulf based off of Red Hat Linux 7.3 and some > other modifications. I'm using BProc and have been suing Clustermatic > as an example and to get some of the bits I need. I've got my patched > kernel, BProc, and beoboot installed but I'm running into some issues > attempting to create phase 1 boot disks with beoboot. I'm getting lots > of unresolved symbols and such when linking together init.o and stubs.o > in the create_init() function. Attached is what I seen on screen. Any > suggestions on how to fix this? > > Also, is there more up to date software for BeoNSS? The only places I > can find it are in Clustermatic and Scyld's FTP site. > > Thanks, > Jack Neely > > -- > Jack Neely <sl...@qu...> > Linux Realm Kit Administration and Development > PAMS Computer Operations at NC State University > GPG Fingerprint: 1917 5AC1 E828 9337 7AA4 EA6B 213B 765F 3B6A 5B89 > # beoboot -1 -f -k /boot//vmlinuz-2.4.19-ncsubeo.0beoboot > Building phase 1 file system image in /tmp/beoboot.2489... > /usr/lib/gcc-lib/i386-redhat-linux/2.96/../../../libmodutils.a(rmmod.o): In function `find_insmod_symbol': > rmmod.o(.text+0x2f): undefined reference to `n_module_stat' > rmmod.o(.text+0x43): undefined reference to `module_stat' > rmmod.o(.text+0x76): undefined reference to `symprefix' > /usr/lib/gcc-lib/i386-redhat-linux/2.96/../../../libmodutils.a(rmmod.o): In function `find_insmod_sections': > rmmod.o(.text+0x14a): undefined reference to `module_stat' > rmmod.o(.text+0x150): undefined reference to `n_module_stat' > rmmod.o(.text+0x192): undefined reference to `symprefix' > rmmod.o(.text+0x306): undefined reference to `n_module_stat' > /usr/lib/gcc-lib/i386-redhat-linux/2.96/../../../libmodutils.a(rmmod.o): In function `rmmod_main': > rmmod.o(.text+0x349): undefined reference to `error_file' > rmmod.o(.text+0x352): undefined reference to `get_kernel_info' > rmmod.o(.text+0x359): undefined reference to `l_module_name_list' > rmmod.o(.text+0x36a): undefined reference to `module_name_list' > rmmod.o(.text+0x370): undefined reference to `l_module_name_list' > rmmod.o(.text+0x3e1): undefined reference to `setsyslog' > rmmod.o(.text+0x3ee): undefined reference to `flag_verbose' > rmmod.o(.text+0x4d6): undefined reference to `get_kernel_info' > rmmod.o(.text+0x51e): undefined reference to `module_stat' > rmmod.o(.text+0x52e): undefined reference to `n_module_stat' > rmmod.o(.text+0x553): undefined reference to `n_module_stat' > rmmod.o(.text+0x582): undefined reference to `n_module_stat' > rmmod.o(.text+0x588): undefined reference to `module_stat' > rmmod.o(.text+0x629): undefined reference to `n_module_stat' > /usr/lib/gcc-lib/i386-redhat-linux/2.96/../../../libmodutils.a(combined_insmod.o): In function `get_kernel_version': > combined_insmod.o(.text+0xd): undefined reference to `uts_info' > combined_insmod.o(.text+0x22): undefined reference to `uts_info' > /usr/lib/gcc-lib/i386-redhat-linux/2.96/../../../libmodutils.a(combined_insmod.o): In function `set_ncv_prefix': > combined_insmod.o(.text+0xb7): undefined reference to `ksyms' > combined_insmod.o(.text+0xbd): undefined reference to `nksyms' > combined_insmod.o(.text+0x19d): undefined reference to `nksyms' > combined_insmod.o(.text+0x1b5): undefined reference to `uts_info' > combined_insmod.o(.text+0x217): undefined reference to `flag_verbose' > combined_insmod.o(.text+0x22d): undefined reference to `lprintf' > /usr/lib/gcc-lib/i386-redhat-linux/2.96/../../../libmodutils.a(combined_insmod.o): In function `ncv_symbol_hash': > combined_insmod.o(.text+0x384): undefined reference to `obj_elf_hash_n' > /usr/lib/gcc-lib/i386-redhat-linux/2.96/../../../libmodutils.a(combined_insmod.o): In function `add_symbols_from': > combined_insmod.o(.text+0x3dc): undefined reference to `obj_find_symbol' > combined_insmod.o(.text+0x404): undefined reference to `obj_add_symbol' > /usr/lib/gcc-lib/i386-redhat-linux/2.96/../../../libmodutils.a(combined_insmod.o): In function `add_kernel_symbols': > combined_insmod.o(.text+0x43b): undefined reference to `n_module_stat' > combined_insmod.o(.text+0x441): undefined reference to `module_stat' > combined_insmod.o(.text+0x47e): undefined reference to `n_module_stat' > combined_insmod.o(.text+0x485): undefined reference to `nksyms' > combined_insmod.o(.text+0x49c): undefined reference to `ksyms' > /usr/lib/gcc-lib/i386-redhat-linux/2.96/../../../libmodutils.a(combined_insmod.o): In function `hide_special_symbols': > combined_insmod.o(.text+0x4d6): undefined reference to `obj_find_symbol' > /usr/lib/gcc-lib/i386-redhat-linux/2.96/../../../libmodutils.a(combined_insmod.o): In function `get_modinfo_value': > combined_insmod.o(.text+0x517): undefined reference to `obj_find_section' > /usr/lib/gcc-lib/i386-redhat-linux/2.96/../../../libmodutils.a(combined_insmod.o): In function `create_this_module': > combined_insmod.o(.text+0x5c7): undefined reference to `obj_create_alloced_section_first' > combined_insmod.o(.text+0x5ef): undefined reference to `obj_add_symbol' > combined_insmod.o(.text+0x600): undefined reference to `obj_string_patch' > /usr/lib/gcc-lib/i386-redhat-linux/2.96/../../../libmodutils.a(combined_insmod.o): In function `add_ksymtab': > combined_insmod.o(.text+0x62d): undefined reference to `obj_find_section' > combined_insmod.o(.text+0x65c): undefined reference to `obj_create_alloced_section' > combined_insmod.o(.text+0x679): undefined reference to `obj_symbol_patch' > combined_insmod.o(.text+0x68c): undefined reference to `obj_string_patch' > /usr/lib/gcc-lib/i386-redhat-linux/2.96/../../../libmodutils.a(combined_insmod.o): In function `create_module_ksymtab': > combined_insmod.o(.text+0x6e1): undefined reference to `obj_create_alloced_section' > combined_insmod.o(.text+0x705): undefined reference to `obj_find_symbol' > combined_insmod.o(.text+0x711): undefined reference to `n_module_stat' > combined_insmod.o(.text+0x725): undefined reference to `module_stat' > combined_insmod.o(.text+0x74c): undefined reference to `obj_symbol_patch' > combined_insmod.o(.text+0x764): undefined reference to `n_module_stat' > combined_insmod.o(.text+0x782): undefined reference to `obj_find_section' > /usr/lib/gcc-lib/i386-redhat-linux/2.96/../../../libmodutils.a(combined_insmod.o): In function `get_module_version': > combined_insmod.o(.text+0x85b): undefined reference to `obj_find_symbol' > combined_insmod.o(.text+0x871): undefined reference to `obj_find_symbol' > /usr/lib/gcc-lib/i386-redhat-linux/2.96/../../../libmodutils.a(combined_insmod.o): In function `is_kernel_checksummed': > combined_insmod.o(.text+0x929): undefined reference to `ksyms' > combined_insmod.o(.text+0x933): undefined reference to `nksyms' > /usr/lib/gcc-lib/i386-redhat-linux/2.96/../../../libmodutils.a(combined_insmod.o): In function `is_module_checksummed': > combined_insmod.o(.text+0x9b9): undefined reference to `obj_find_symbol' > /usr/lib/gcc-lib/i386-redhat-linux/2.96/../../../libmodutils.a(combined_insmod.o): In function `add_ksymoops_symbols': > combined_insmod.o(.text+0xa67): undefined reference to `obj_find_section' > combined_insmod.o(.text+0xa90): undefined reference to `obj_find_section' > combined_insmod.o(.text+0xafe): undefined reference to `symprefix' > combined_insmod.o(.text+0xb22): undefined reference to `obj_add_symbol' > combined_insmod.o(.text+0xb6f): undefined reference to `obj_find_section' > combined_insmod.o(.text+0xbaf): undefined reference to `symprefix' > combined_insmod.o(.text+0xbd5): undefined reference to `obj_add_symbol' > /usr/lib/gcc-lib/i386-redhat-linux/2.96/../../../libmodutils.a(combined_insmod.o): In function `process_module_arguments': > combined_insmod.o(.text+0xcc2): undefined reference to `flag_verbose' > combined_insmod.o(.text+0xcda): undefined reference to `lprintf' > combined_insmod.o(.text+0xd96): undefined reference to `obj_find_symbol' > combined_insmod.o(.text+0xf66): undefined reference to `obj_string_patch' > /usr/lib/gcc-lib/i386-redhat-linux/2.96/../../../libmodutils.a(combined_insmod.o): In function `add_kallsyms': > combined_insmod.o(.text+0x1132): undefined reference to `ksyms' > combined_insmod.o(.text+0x1141): undefined reference to `nksyms' > combined_insmod.o(.text+0x1255): undefined reference to `obj_create_alloced_section' > combined_insmod.o(.text+0x126b): undefined reference to `obj_kallsyms' > combined_insmod.o(.text+0x12bc): undefined reference to `obj_free' > /usr/lib/gcc-lib/i386-redhat-linux/2.96/../../../libmodutils.a(combined_insmod.o): In function `add_archdata': > combined_insmod.o(.text+0x1365): undefined reference to `obj_create_alloced_section' > combined_insmod.o(.text+0x137c): undefined reference to `obj_arch_archdata' > /usr/lib/gcc-lib/i386-redhat-linux/2.96/../../../libmodutils.a(combined_insmod.o): In function `init_module': > combined_insmod.o(.text+0x13ae): undefined reference to `obj_find_section' > combined_insmod.o(.text+0x13c9): undefined reference to `flag_autoclean' > combined_insmod.o(.text+0x13e1): undefined reference to `obj_find_section' > combined_insmod.o(.text+0x1413): undefined reference to `obj_find_section' > combined_insmod.o(.text+0x1433): undefined reference to `obj_find_symbol' > combined_insmod.o(.text+0x143c): undefined reference to `obj_symbol_final_value' > combined_insmod.o(.text+0x144c): undefined reference to `obj_find_symbol' > combined_insmod.o(.text+0x1455): undefined reference to `obj_symbol_final_value' > combined_insmod.o(.text+0x1465): undefined reference to `obj_find_section' > combined_insmod.o(.text+0x1487): undefined reference to `obj_find_section' > combined_insmod.o(.text+0x14a5): undefined reference to `obj_find_section' > combined_insmod.o(.text+0x14d3): undefined reference to `obj_find_section' > combined_insmod.o(.text+0x14fb): undefined reference to `obj_find_section' > combined_insmod.o(.text+0x151f): undefined reference to `obj_arch_init_module' > combined_insmod.o(.text+0x1546): undefined reference to `obj_create_image' > combined_insmod.o(.text+0x15e5): undefined reference to `sys_init_module' > combined_insmod.o(.text+0x1608): undefined reference to `lprintf' > /usr/lib/gcc-lib/i386-redhat-linux/2.96/../../../libmodutils.a(combined_insmod.o): In function `check_module_parameter': > combined_insmod.o(.text+0x1644): undefined reference to `obj_find_symbol' > combined_insmod.o(.text+0x1653): undefined reference to `error_file' > combined_insmod.o(.text+0x16ba): undefined reference to `error_file' > combined_insmod.o(.text+0x1718): undefined reference to `error_file' > combined_insmod.o(.text+0x1748): undefined reference to `error_file' > combined_insmod.o(.text+0x175b): undefined reference to `error_file' > /usr/lib/gcc-lib/i386-redhat-linux/2.96/../../../libmodutils.a(combined_insmod.o)(.text+0x177f): more undefined references to `error_file' follow > /usr/lib/gcc-lib/i386-redhat-linux/2.96/../../../libmodutils.a(combined_insmod.o): In function `check_module_parameter': > combined_insmod.o(.text+0x1789): undefined reference to `lprintf' > /usr/lib/gcc-lib/i386-redhat-linux/2.96/../../../libmodutils.a(combined_insmod.o): In function `check_module_parameters': > combined_insmod.o(.text+0x17b9): undefined reference to `obj_find_section' > /usr/lib/gcc-lib/i386-redhat-linux/2.96/../../../libmodutils.a(combined_insmod.o): In function `check_tainted_module': > combined_insmod.o(.text+0x19e5): undefined reference to `obj_gpl_license' > /usr/lib/gcc-lib/i386-redhat-linux/2.96/../../../libmodutils.a(combined_insmod.o): In function `insmod_main': > combined_insmod.o(.text+0x1b15): undefined reference to `error_file' > combined_insmod.o(.text+0x1b29): undefined reference to `errors' > combined_insmod.o(.text+0x1b6a): undefined reference to `flag_autoclean' > combined_insmod.o(.text+0x1bac): undefined reference to `root_check_off' > combined_insmod.o(.text+0x1bb5): undefined reference to `root_check_off' > combined_insmod.o(.text+0x1bc5): undefined reference to `setsyslog' > combined_insmod.o(.text+0x1bd2): undefined reference to `flag_verbose' > combined_insmod.o(.text+0x1c8b): undefined reference to `config_read' > combined_insmod.o(.text+0x1d7a): undefined reference to `search_module_path' > combined_insmod.o(.text+0x1da6): undefined reference to `flag_verbose' > combined_insmod.o(.text+0x1dbb): undefined reference to `lprintf' > combined_insmod.o(.text+0x1e1d): undefined reference to `get_kernel_info' > combined_insmod.o(.text+0x1e46): undefined reference to `n_module_stat' > combined_insmod.o(.text+0x1e5d): undefined reference to `module_stat' > combined_insmod.o(.text+0x1e97): undefined reference to `error_file' > combined_insmod.o(.text+0x1e9c): undefined reference to `obj_load' > combined_insmod.o(.text+0x1f2e): undefined reference to `lprintf' > combined_insmod.o(.text+0x1f73): undefined reference to `obj_set_symbol_compare' > combined_insmod.o(.text+0x1f80): undefined reference to `obj_gpl_license' > combined_insmod.o(.text+0x1fb2): undefined reference to `obj_arch_create_got' > combined_insmod.o(.text+0x1fc0): undefined reference to `obj_check_undefineds' > combined_insmod.o(.text+0x200d): undefined reference to `obj_allocate_commons' > combined_insmod.o(.text+0x2089): undefined reference to `k_new_syscalls' > combined_insmod.o(.text+0x20d9): undefined reference to `errors' > combined_insmod.o(.text+0x20f6): undefined reference to `obj_load_size' > combined_insmod.o(.text+0x2181): undefined reference to `obj_relocate' > combined_insmod.o(.text+0x21d8): undefined reference to `errors' > /usr/lib/gcc-lib/i386-redhat-linux/2.96/../../../libmodutils.a(combined_insmod.o): In function `add_ksymtab': > combined_insmod.o(.text+0x6a5): undefined reference to `obj_extend_section' > collect2: ld returned 1 exit status > beoboot: Fatal Error (aborting): Failed to build init for ramdisk image. ----[ Wilton William Wong ]--------------------------------------------- 11060-166 Avenue Ph : 01-780-456-9771 High Performance UNIX Edmonton, Alberta FAX: 01-780-456-9772 and Linux Solutions T5X 1Y3, Canada URL: http://www.harddata.com -------------------------------------------------------[ Hard Data Ltd. ]---- |
|
From: Jack N. <jj...@pa...> - 2002-09-16 19:20:03
|
Hi, I'm trying to create a Beowulf based off of Red Hat Linux 7.3 and some other modifications. I'm using BProc and have been suing Clustermatic as an example and to get some of the bits I need. I've got my patched kernel, BProc, and beoboot installed but I'm running into some issues attempting to create phase 1 boot disks with beoboot. I'm getting lots of unresolved symbols and such when linking together init.o and stubs.o in the create_init() function. Attached is what I seen on screen. Any suggestions on how to fix this? Also, is there more up to date software for BeoNSS? The only places I can find it are in Clustermatic and Scyld's FTP site. Thanks, Jack Neely -- Jack Neely <sl...@qu...> Linux Realm Kit Administration and Development PAMS Computer Operations at NC State University GPG Fingerprint: 1917 5AC1 E828 9337 7AA4 EA6B 213B 765F 3B6A 5B89 |
|
From: Larry B. <ba...@us...> - 2002-09-10 22:03:03
|
I have a working master/2-slave node Clustermatic, all using Intel P4 = "A"'s. For testing cluster changes, I slide in an Intel PII in place of = the master. I just received 2 1 GHz AMD Athlons that I would like to = use with my PII for testing cluster changes. What's the smartest way to = install Clustermatic twice so that the PII master runs the i686 code and = the Athlon slaves run the athlon code, assuming that is required? Larry Baker US Geological Survey |
|
From: <er...@he...> - 2002-09-03 19:22:46
|
On Sun, Sep 01, 2002 at 08:05:04AM -0700, Roberto Aguiar Veiga wrote: > > Hello, friends: > > I'm responsible for implementing a cluster of Linux machines at the Department of Physics at the Federal University of Uberlandia (Brazil). Surfing through the web, I discovered BProc and I am seriously thinking about using it. > > I have performed tests with Mosix. I'd like to know: are Mosix and BProc similar? Is BProc by itself capable of deal with parallel applications or will I need other softwares, like MPI or PVM? BProc is not that similar to Mosix. BProc does not do automatic or transparent process migration. The good thing about it is that there's no added overhead for most applications when running on remote machines. You will still need MPI or some other library to do parallel computation. - Erik |
|
From: Roberto A. V. <ra...@ya...> - 2002-09-01 15:05:10
|
Hello, friends:
I'm responsible for implementing a cluster of Linux machines at the Department of Physics at the Federal University of Uberlandia (Brazil). Surfing through the web, I discovered BProc and I am seriously thinking about using it.
I have performed tests with Mosix. I'd like to know: are Mosix and BProc similar? Is BProc by itself capable of deal with parallel applications or will I need other softwares, like MPI or PVM?
Regards,
Roberto Veiga
---------------------------------
Do You Yahoo!?
Yahoo! Finance - Get real-time stock quotes |
|
From: Erik A. H. <er...@he...> - 2002-08-26 20:17:44
|
On Mon, Aug 26, 2002 at 03:32:02PM +0000, Sam Przyswa wrote: > Hi, > > Does BProc is able to run and migrate multithreaded and shared memory apps as > MySQL, Apache ? Running them should be fine unless they use CLONE_THREAD. I don't think anything you've mentioned does that. There's currently no support for migrating multi-threaded processes or processes which share memory. - Erik |
|
From: Sam P. <sa...@ar...> - 2002-08-26 15:40:51
|
Hi, Does BProc is able to run and migrate multithreaded and shared memory apps as MySQL, Apache ? Thanks in advance. Sam. -- Sam Przyswa - Chef de projet Arial Concept - Intégrateur Internet 36, rue de Turin - 75008 - Paris Tel: 01 40 54 86 04 - Fax: 01 40 54 83 01 Web: http://www.arial-concept.com - Email: In...@ar... |
|
From: Shaun K. <ske...@ya...> - 2002-08-22 23:59:57
|
Problem: When send beoserv the SIGHUP signal, the client comes up with an error message: bpslave: libcache: fail list reset I reboot the machine with beoserv reloaded, and I still get the same librt problem. I've installed the libs from the RPM's on the cd, and only include those in the bplib list, but I still don't get any results. Any ideas? -Shaun Keenan CSU Hayward --- Erik Arjan Hendriks <er...@he...> wrote: > On Wed, Aug 21, 2002 at 05:43:24PM -0700, Shaun Keenan wrote: > > Running ClusterMatic, and I can't do much of anything because the slaves > can't > > find any libraries whatsoever. I run setup_libs and it says it can't even > run > > tar on the slave nodes because librt.so.1 is missing. I did a bplib -a > > /lib/librt.so.1 and I still get the same error message. > > You need to place the real file on the library list, not the symlink. > i.e. put /lib/librt-2.2.4.so (or whatever version you have) instead of > /lib/librt.so.1. Then, if you don't reboot the slave in between, make > sure you send a SIGHUP to beoserv. That will tell it to re-read the > list of libraries to distribute. > > > I can't run beofdisk either, node_up doesn't work because there's no > > libs... > > The above problem should fix that. > > - Erik __________________________________________________ Do You Yahoo!? HotJobs - Search Thousands of New Jobs http://www.hotjobs.com |
|
From: Erik A. H. <er...@he...> - 2002-08-22 13:03:44
|
On Wed, Aug 21, 2002 at 05:43:24PM -0700, Shaun Keenan wrote: > Running ClusterMatic, and I can't do much of anything because the slaves can't > find any libraries whatsoever. I run setup_libs and it says it can't even run > tar on the slave nodes because librt.so.1 is missing. I did a bplib -a > /lib/librt.so.1 and I still get the same error message. You need to place the real file on the library list, not the symlink. i.e. put /lib/librt-2.2.4.so (or whatever version you have) instead of /lib/librt.so.1. Then, if you don't reboot the slave in between, make sure you send a SIGHUP to beoserv. That will tell it to re-read the list of libraries to distribute. > I can't run beofdisk either, node_up doesn't work because there's no > libs... The above problem should fix that. - Erik |
|
From: Erik A. H. <er...@he...> - 2002-08-22 12:59:34
|
On Wed, Aug 21, 2002 at 01:29:38PM -0700, Shaun Keenan wrote: > Hey, > I've set up a 3-computer cluster for testing/playing around. Does anyone know > of any good programs to test out the cluster with, or even any opensource > database/web/app servers I could use? We (clustermatic) don't have anything in the way of fun apps. We presume users come with apps :) You might want to try building MPICH and finding some MPI code. > I would like to see what all I can do with this now that I have it > set up. Using ClusterMatic by the way. Also, how much better would > the free version of Scyld be? I have no idea what Scyld is distributing these days but if it's still BProc 2.x based, the BProc with clustermatic (3.x) is going to be much less buggy and more featureful. - Erik |
|
From: Shaun K. <ske...@ya...> - 2002-08-22 00:43:25
|
Running ClusterMatic, and I can't do much of anything because the slaves can't find any libraries whatsoever. I run setup_libs and it says it can't even run tar on the slave nodes because librt.so.1 is missing. I did a bplib -a /lib/librt.so.1 and I still get the same error message. I can't run beofdisk either, node_up doesn't work because there's no libs... any suggestions? -Shaun Keenan CSU Hayward __________________________________________________ Do You Yahoo!? HotJobs - Search Thousands of New Jobs http://www.hotjobs.com |
|
From: Shaun K. <ske...@ya...> - 2002-08-21 20:29:39
|
Hey, I've set up a 3-computer cluster for testing/playing around. Does anyone know of any good programs to test out the cluster with, or even any opensource database/web/app servers I could use? I would like to see what all I can do with this now that I have it set up. Using ClusterMatic by the way. Also, how much better would the free version of Scyld be? -Shaun Keenan CSU Hayward __________________________________________________ Do You Yahoo!? HotJobs - Search Thousands of New Jobs http://www.hotjobs.com |
|
From: Erik A. H. <er...@he...> - 2002-08-21 15:22:14
|
On Wed, Aug 21, 2002 at 10:23:00AM -0400, Nicholas Henke wrote:
> I have made the following fix to the bproc.spec file. This allows bproc to
> be installed on a machine with only the smp kernel installed.
>
> --- SPECS/bproc-orig.spec Wed Aug 21 10:21:20 2002
> +++ SPECS/bproc.spec Wed Aug 21 10:21:46 2002
> @@ -80,6 +80,7 @@
> Group: System Environment/Libraries
> Release: 1%{releaseextra}
> Requires: kernel-smp = %{kernelvers}
> +Provides: bproc-modules
>
> %description modules-smp
> The BProc distributed process space is a central process
Thanks, applied.
- Erik
|
|
From: Nicholas H. <he...@se...> - 2002-08-21 14:23:22
|
I have made the following fix to the bproc.spec file. This allows bproc to
be installed on a machine with only the smp kernel installed.
--- SPECS/bproc-orig.spec Wed Aug 21 10:21:20 2002
+++ SPECS/bproc.spec Wed Aug 21 10:21:46 2002
@@ -80,6 +80,7 @@
Group: System Environment/Libraries
Release: 1%{releaseextra}
Requires: kernel-smp = %{kernelvers}
+Provides: bproc-modules
%description modules-smp
The BProc distributed process space is a central process
Nic
--
Nicholas Henke
Linux Cluster Systems Programmer
he...@se... - 215.573.8149
|
|
From: Erik A. H. <er...@he...> - 2002-08-15 21:33:27
|
On Wed, Aug 14, 2002 at 04:51:08PM -0600, David wrote: > When trying to send a command to a node using bpsh, I get the following error: > bproc_vexecmove_io: Invalid argument > > Using the -N option (preventing io redirection) works fine, and the command goes through. > Ok, you caught me. I didn't test this release on PPC before releasing it. I'll try it out when I can borrow a Mac a again. Until then though, here are a couple of things to check. The vexecmove interface (API and kernel) changed a bit since the last release. There is a possibility that you'll get an old library revision if you didn't run ldconfig after installing. In any case, I'd check the symlinks in /usr/lib and "ldd bpsh" to make sure it's getting the new shared library. - Erik |
|
From: David <dg...@te...> - 2002-08-14 22:51:09
|
When trying to send a command to a node using bpsh, I get the following e=
rror:
bproc_vexecmove_io: Invalid argument
Using the -N option (preventing io redirection) works fine, and the comma=
nd goes through.
The strace is below:
SYS_223(0x201, 0x10014d80, 0x40, 0x10014d78, 0) =3D 1
socket(PF_INET, SOCK_STREAM, IPPROTO_IP) =3D 3
bind(3, {sin_family=3DAF_INET, sin_port=3Dhtons(0), sin_addr=3Dinet_addr(=
"0.0.0.0")}}, 16) =3D 0
listen(3, 1024) =3D 0
getsockname(3, {sin_family=3DAF_INET, sin_port=3Dhtons(32849), sin_addr=3D=
inet_addr("0.0.0.0")}}, [16]) =3D 0
fcntl64(3, F_GETFL) =3D 0x2 (flags O_RDWR)
fcntl64(3, F_SETFL, 0x4) =3D 0
fcntl64(3, F_GETFL) =3D 0x2 (flags O_RDWR)
fcntl64(3, F_SETFL, O_RDWR|O_NONBLOCK) =3D 0
SYS_223(0x201, 0x7ffff378, 0x1, 0x4, 0x33) =3D 1
brk(0x10019000) =3D 0x10019000
clone(child_stack=3D0x100172e0, flags=3DCLONE_VM|CLONE_FS|CLONE_FILES) =3D=
3109
SYS_223(0x309, 0x7ffff338, 0x7ffff3c8, 0x3, 0x10014d68) =3D -1 EINVAL (In=
valid argument)
write(2, "bproc_vexecmove_io: Invalid argu"..., 37bproc_vexecmove_io: Inv=
alid argument
) =3D 37
kill(3109, SIGTERM) =3D 0
wait4(3109, NULL, __WCLONE, NULL) =3D 3109
exit(1) =3D ?
Anyone?
|
1488 messages has been excluded from this view by a project administrator.
