Hi Predrag!
Predrag Punosevac <[email protected]> wrote:
|predrag@oko$ uname -a
|OpenBSD oko.bagdala2.net 5.9 GENERIC.MP#1 amd64
Hm, not today, i'd need to go to the city:
mreget install59.iso
mreget install59.iso [anpqy?]? y
ftp: Can't stat `install59.iso': No such file or directory
229 Extended Passive mode OK (|||55958|)
150-Accepted data connection
150 217910.0 kbytes to download
5% |***** | 11462 KiB 30.56 KiB/s
1:52:33 ETA
10% |*********** | 23523 KiB 26.19 KiB/s
2:03:40 ETA
14% |**************** | 32143 KiB 19.78 KiB/s
2:36:31 ETA
31% |*********************************** | 68861 KiB 15.35 KiB/s -
stalled -
33% |************************************** | 73312 KiB 14.52 KiB/s
2:45:56 ETA
|predrag@oko$ s-nail
|s-nail version v14.8.8. Type ? for help.
?1[steffen@obsd s-nail-14.8.8]$ uname -a
OpenBSD obsd.my.domain 5.8 GENERIC#1276 i386
|predrag@oko$ egdb
Ugh! No!! I'm sorry for you! A debugger! Yuck!!
|(gdb) core-file s-nail.core
|[New process 4875]
|Core was generated by `s-nail'.
|Program terminated with signal SIGBUS, Bus error.
|#0 0x0000061eff12788b in ?? ()
Oh yes, thanks for the informative bug report.
It clearly shows that you should have installed the program with
root privileges so that the dotlock program can be installed
SETUID which it needs for working properly (especially on OpenBSD).
E.g., you shouldn't have kill(1)ed it, just wait some more
seconds, then you should be able to read
Creating dotlock for "/var/mail/steffen" .....
Unable to (dot) lock mailbox, aborting operation: Permission denied
If you then run S-nail again the message should appear faster.
The manual has an actual entry
FAQ
S-nail shortly hangs on startup
And that makes me wonder wether you have oko.bagdala2.net in
/etc/hosts or something. I don't know why a BSD should hang in
uname(2) though, and i don't have no sources on this box (and
i also don't have the OpenBSD sources no more anyway since the git
clone has been deserted). In fact i never really understood that,
but also never really was interested enough to dig into it.
I see now however that the signals are blocked when uname(2) (may)
get called, which seems to hurt especially the first time.
I cannot simply change this, i will have to implement a global
state that all facilities that possibly end up looking into the
nodename check before diving deeper in the code. I guess that is
why i didn't implement it at first glance. But i think i have to.
In general it is always a good idea to run S-nail with -d and/or
-vv when problems happen. But this also can improve for sure.
Does this answer your question? ^_^
Thanks for the report.
Ciao!
--steffen
------------------------------------------------------------------------------
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
__________________________________
[email protected]
