Valgrind: Killed (FC5)
by wwp
Hi there,
I'm trying to use valgrind in FC5..
$ valgrind --help
Killed
Any other attempt to run valgrind w/ a correct command-line gives the same
error (user or root). I tried valgrind-3.1.0-2 or 3.1.1 from the sources.
FYI, kernel is 2.6.16-1.2096_FC5.custom3 (stripped down to me smaller towards
hardware drivers), no se_linux.
Any hint? Did I miss anything?
strace output:
$ strace valgrind --help
execve("/usr/bin/valgrind", ["valgrind", "--help"], [/* 40 vars */]) = 0
brk(0) = 0x9bd7000
mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0xa7f76000
access("/etc/ld.so.preload", R_OK) = -1 ENOENT (No such file or directory)
open("/usr/local/lib/tls/i686/sse2/libc.so.6", O_RDONLY) = -1 ENOENT (No such file or directory)
open("/usr/local/lib/tls/i686/libc.so.6", O_RDONLY) = -1 ENOENT (No such file or directory)
open("/usr/local/lib/tls/sse2/libc.so.6", O_RDONLY) = -1 ENOENT (No such file or directory)
open("/usr/local/lib/tls/libc.so.6", O_RDONLY) = -1 ENOENT (No such file or directory)
open("/usr/local/lib/i686/sse2/libc.so.6", O_RDONLY) = -1 ENOENT (No such file or directory)
open("/usr/local/lib/i686/libc.so.6", O_RDONLY) = -1 ENOENT (No such file or directory)
open("/usr/local/lib/sse2/libc.so.6", O_RDONLY) = -1 ENOENT (No such file or directory)
open("/usr/local/lib/libc.so.6", O_RDONLY) = -1 ENOENT (No such file or directory)
open("/etc/ld.so.cache", O_RDONLY) = 4
fstat64(4, {st_mode=S_IFREG|0644, st_size=121122, ...}) = 0
mmap2(NULL, 121122, PROT_READ, MAP_PRIVATE, 4, 0) = 0xa7f58000
close(4) = 0
open("/lib/libc.so.6", O_RDONLY) = 4
read(4, "\177ELF\1\1\1\0\0\0\0\0\0\0\0\0\3\0\3\0\1\0\0\0J\330c\000"..., 512) = 512
fstat64(4, {st_mode=S_IFREG|0755, st_size=1532536, ...}) = 0
mmap2(0x628000, 1254780, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_DENYWRITE, 4, 0) = 0x628000
mmap2(0x755000, 12288, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 4, 0x12d) = 0x755000
mmap2(0x758000, 9596, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0x758000
close(4) = 0
mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0xa7f57000
set_thread_area({entry_number:-1 -> 6, base_addr:0xa7f576c0, limit:1048575, seg_32bit:1, contents:0, read_exec_only:0, limit_in_pages:1, seg_not_present:0, useable:1}) = 0
mprotect(0x755000, 8192, PROT_READ) = 0
mprotect(0x624000, 4096, PROT_READ) = 0
munmap(0xa7f58000, 121122) = 0
readlink("/proc/self/exe", "/usr/bin/valgrind", 4096) = 17
brk(0) = 0x9bd7000
brk(0x9bf8000) = 0x9bf8000
execve("/usr/lib/valgrind/x86-linux/memcheck", ["valgrind", "--help"], [/* 41 vars */]) = 0
+++ killed by SIGKILL +++
Process 21593 detached
Regards,
--
wwp
17 years, 11 months
FC Pine passfile default? Maintainer(s)?
by Beartooth
Over on comp.mail.pine, I recently lucked into catching the invaluable
interest of Nancy McGough, the authority on making Pine jump through hoops
and like it at www.ii.com. With her vast help, I'm at least halfway (as
near as I can judge) to setting up a master Pine configuration to handle
my very different email accounts from one place; and at this point she has
recommended, and after consulting my local LUG I have agreed, that I would
do well to set up a passfile.
But there we hit a snag. She asks :
> OK, let's go for it! I can walk you through setting it up using
> the '-passfile' command-line argument, but it would be useful for
> you (and others!) to find out the default value that this RedHat
> Pine uses. Could you please ask the maintainer of that RH Pine
> package what is the default Pine PASSFILE that he/she built this
> RH Pine with? That way I can write about it in the Pine docs I'm
> writing.
The snag, of course (as I have just explained, I hope not too verbosely,
on c.m.p) is that I have no idea who the maintainer is. (Incidentally, it
would be a vast benefit if someone could find a way to make "yum install
pine" work, presumably through some one or more of the unofficial repos.)
My own procedure, fwiw, is to get an rpm for each new Fedora, from UW or
Dag or livna; try "rpm -ivh" (or -Uvh); get a dependency or three (nearly
always); command "yum install <dependency>" -- hoping for all I'm worth to
see that succeed; repeat the rpm command; and turn instantly back into a
happy camper.
Until all that works, I keep one machine un-upgraded, just for running
Pine; I'm a diehard's diehard about that. (I took up linux, years ago,
most of all in order to be able to run real Pine.)
If there at least *IS* a Fedora default passfile for Pine, common to
maintainers of Pine for Fedora, what is it? Or if there is one maintainer,
or a chief maintainer, who is it and is there somewhere else I should ask?
If there are two or three defaults (for UW, livna, and Dag perhaps?), what
are they?
Fwiw, I'm sure my own present instance is from livna; but the more
definitively and completely I can report back to comp.mail.pine, and for
Nancy's splendid Pine FAQ, the better.
--
Beartooth Staffwright, Wordcrafty Squirreler, Neo-
Redneck Retiree, Not Quite Clueless FC Power User
17 years, 11 months
lvm boot from second volume group
by Mike Wright
Hi everybody!
I'm using fc4 and working with xen, and it works great, but I've come
upon an issue that I don't know how to resolve.
I've many different images and configurations using lvm and grub to
multi-boot. To keep things easy (for me) I use IDE drives and the
motherboard's built-in IDE interfaces.
I added another volume group, vg1, to my lvm setup, and placed an fc4
tree on it, and added an entry to grub.conf. However, when I try to
boot into that image it fails. Same image on volume group 0 (vg0) will
boot.
Has anybody managed to use grub to boot from a second volume group?
I suspect that it may be an initrd issue, but don't know for sure.
Any pointers or suggestions would be appreciated.
Thanx!
17 years, 11 months
system clock misbehaving
by Ali Sobhi
I've been struggling with system clock on my system since day 1 that I
installed FC5.
I did not have any problems with FC4.
I stopped using NPTD for time sync, turned off most of the services, tried
to look through
log files and can't find anything which tells me what is happening. I have
not turned off any cron or anacron activities yet.
Here is what I'm seeing with regard to system clock. After about 48 to 56
hours of up time, the system slows down,
no apparent memory or CPU hog on the queue. The response time goes up for
a simple date or ls command.
looking at "/proc/driver/rtc" and date output, there is about 8+ hours
difference which normally happens within very
short period of time (~ 4 hours). Continuously typing date, I can see the
clock going back and forth. "sleep 1" takes about couple of
minutes to complete. My only solution so far has been to do the following
twice and then "sync; reboot -f"
ntpdate <MyTimeServer>
rm -f /etc/adjtime
hwclock -w
After this, everything seems OK until another 48 to 56 hours. I'm thinking
about putting this in a cron job and rebooting the machine
every morning.
I have to mention that I have two different type of machines that have
been up and running for 32 days now without any time problem and both
are using NTPD.
any pointers to where to look for source of problem and/or additional info
would be helpful.
Regards,
Ali Sobhi
-------------------------------------------------------------------------
Sr. Consultant - IBM Research
Human Ability and Accessibility Center
512-823-0064 (T/L 793) sobhi(a)us.ibm.com
http://www.ibm.com/able/
-------------------------------------------------------------------------
17 years, 11 months
fc5 firefox won't open pdf: Unhandled MIME type: 'application/octet-stream'
by sean darcy
Since upgrading to fc5 firefox won't show pdf docs. I can
save the doc and open it in acroread, evince, etc.
When firefox tries to to show a pdf file I get an error window:
Unable to open document
Unhandled MIME type: 'application/octet-stream'
I've messed with mozpluggerrc to use acroread, then I tried
evince. Same problem.
FWIW, x86_64.
Any help appreciated
sean
17 years, 11 months
the safety of gnupg
by Tim
I've just been reading some rather silly things about gnupg except for
one practical point: Who has actually checked the source code for it to
see whether it's trustworthy, etc?
And, of course, the next thing would be: Who would they be that we
could trust them, too? After a bit of Googling around, I'm darned if I
can find out, nor think of the right terms to search for.
Naturally, this extends to other open-source programs. Which ones do
get vetted?
--
(Currently running FC4, occasionally trying FC5.)
Don't send private replies to my address, the mailbox is ignored.
I read messages from the public lists.
17 years, 11 months
/dev/tape
by Les Mikesell
After installing FC5, I don't see /dev/tape showing up as a symlink
to /dev/nst0 anymore and it disappears after a reboot if I add one
manually to make mt and amanda work. The tape drive wasn't attached
when I installed - would that make a difference?
/etc/udev/rules.d/50-udev.rules does have an entry:
KERNEL=="nst[0-9]", BUS=="scsi", SYMLINK+="tape-%k", MODE="0660"
and I do have /dev/nst0. Does it need something else?
--
Les Mikesell
lesmikesell(a)gmail.com
17 years, 11 months
FC5 & IBM xSeries 100
by Jouni Viikari
Hi all,
I have installed FC5 to an brand new IBM xSeries 100 type 8486 server.
On short trials It seems to work almost normally except:
1) On boot I always get following messages on console:
PCI: Cannot allocate resource region 7 of bridge 0000:00:01.0
PCI: Cannot allocate resource region 8 of bridge 0000:00:01.0
PCI: Cannot allocate resource region 9 of bridge 0000:00:01.0
PCI: Cannot allocate resource region 7 of bridge 0000:00:1c.1
PCI: Cannot allocate resource region 8 of bridge 0000:00:1c.1
PCI: Cannot allocate resource region 9 of bridge 0000:00:1c.1
2) On short trial the server seems to work quite normally except when I
try to start X. startx (or gdm at runlevel 5) freezes most of the times
(not always) and leaves screen blank and in unusable state. I still can
login to the box through ssh (Server works slowly in this state).
Following gets logged during X crash:
kernel: [drm] Initialized drm 1.0.1 20051102
kernel: ACPI: PCI Interrupt 0000:0a:04.0[A] -> GSI 16 (level, low) ->
IRQ 16
kernel: [drm] Initialized radeon 1.22.0 20051229 on minor 0
kernel: irq 20: nobody cared (try booting with the "irqpoll" option)
kernel: [<c0145ac7>] __report_bad_irq+0x2b/0x69 [<c0145c88>]
note_interrupt+0x183/0x1af
kernel: [<c01455f2>] handle_IRQ_event+0x23/0x4c [<c01456c8>]
__do_IRQ+0xad/0xdd
kernel: [<c0105eee>] do_IRQ+0x60/0x7b =======================
kernel: [<c0104786>] common_interrupt+0x1a/0x20
kernel: [<c0102f65>] mwait_idle+0x1f/0x33 [<c0102f2d>] cpu_idle
+0x8f/0xa8
kernel: [<c03c8715>] start_kernel+0x2fe/0x304 <3>handlers:
kernel: [<c026e461>] (usb_hcd_irq+0x0/0x50)
kernel: Disabling IRQ #20
This has happened with every kernel available to FC5 so far.
Since this is a box I promised one person to setup as a linux server and
it has been already sitting here one month unused, I wonder if anybody
knows if it is safe to put it under 'production-use' (If no X is used
and just acting as a pure mail/web server)?
I have tried some boot options (e.g. acpi=off pci=noacpi acpi=noirq, but
on the kernels I tried they did not help).
Added comments on error #186557
BR,
Jouni
17 years, 11 months
Trouble starting postgresql
by Alan M. Evans
Hello!
On my FC5 server, I wanted to configure postgresql to use
/home/pgsql/data instead of /var/lib/pgsql/data. (/home partition is
huge and doesn't get formatted when system is upgraded or new system
installed.) I created the directory, ran initdb; everything seemed ok.
Research seemed to tell me that I should perform the reconfiguration in
/etc/sysconfig/pgsql. However, I could not start the service when the
new directory was specified there. Much experimentation led me to this:
[root@citadel ~]# rm -f /home/pgsql/pgstartup.log
[root@citadel ~]# cp /etc/rc.d/init.d/postgresql
/etc/rc.d/init.d/postgresql2
cp: overwrite `/etc/rc.d/init.d/postgresql2'? y
[root@citadel ~]# cp /etc/sysconfig/pgsql/postgresql
/etc/sysconfig/pgsql/postgresql2
cp: overwrite `/etc/sysconfig/pgsql/postgresql2'? y
[root@citadel ~]# cat /etc/sysconfig/pgsql/postgresql
PGDATA=/home/pgsql/data
PGLOG=/home/pgsql/pgstartup.log
[root@citadel ~]# service postgresql start
Starting postgresql service: [FAILED]
[root@citadel ~]# service postgresql2 start
Starting postgresql2 service: [ OK ]
[root@citadel ~]# service postgresql2 stop
Stopping postgresql2 service: [ OK ]
[root@citadel ~]# cat /home/pgsql/pgstartup.log
postmaster cannot access the server configuration file
"/home/pgsql/data/postgresql.conf": Permission denied
LOG: logger shutting down
[root@citadel ~]#
To sum up: I made an exact copy of the server startup script and an
exact copy of my configuration script. The copy starts, but the original
fails.
Obviously, I'm missing something fundamental here. Any help is
appreciated as I am at the end of ideas.
17 years, 11 months
Error 15: File not found after FC 5 Install
by philbrog
I installed the 5-CD set of Fedora Core 5 and within 5 seconds after I press
Enter on the GRUB menu, it displays these messages:
Booting 'Fedora Core (2.4.22-1.2115.nptl)'
root (hd0,0)
Filesystem is ext2fs, partition type 0x83
kernel /vmlinuz-2.4.22-1.2115.nptl ro root=LABEL=/
Error 15: File not found
Press any key to continue...
Whenever I press any key, it takes me back to the GRUB menu. When I press
Enter from GRUB, it takes me through the error sequence I describe above.
Thanks for your time, I hope you can help.
17 years, 11 months
