Re: Emacsen segfaulting on startup

[Makoto Fujiwara]

My rough memory reminds that Xen kernel generated emacs binary has problem
on regular kernel. Is that problem fixed ?
---
Makoto Fujiwara

Re: [RAIDframe] system hangs while manipulating large files

[Emile `iMil' Heitor]

  FWIW, it is documented in mount(8):


log (FFS only) Mount the file system with wapbl(4) meta-
data journaling, also known simply as logging.  It
provides rapid metadata updates and eliminates the
need to check file system consistency after a system
outage.  A file system mounted with log can not be
mounted with async.  [...]


Well, my bad :(

Wouldn't it be wise that mount(8) forbids the use of both flags ?

--
Emile `iMil' Heitor .°. 
_
  | http://imil.net| ASCII ribbon campaign ( )
  | http://www.NetBSD.org  |  - against HTML email  X
  | http://gcu.info|  & vCards / \

Re: [RAIDframe] system hangs while manipulating large files

[Gary Duzan]

In Message ,
   "Emile `iMil' Heitor" wrote:

=>
=>> asynchronous + log should be impossible.
=>
=>I've just discovered that the hard way :)

   FWIW, it is documented in mount(8):


 log (FFS only) Mount the file system with wapbl(4) meta-
 data journaling, also known simply as logging.  It
 provides rapid metadata updates and eliminates the
 need to check file system consistency after a system
 outage.  A file system mounted with log can not be
 mounted with async.  [...]


Gary Duzan

Re: [RAIDframe] system hangs while manipulating large files

[Emile `iMil' Heitor]

asynchronous + log should be impossible.


I've just discovered that the hard way :)


Did you find out wether the system is really hanging or just very slow?
In particular, is there any disk activity and does the system respond
on the network?


Like I said earlier on this thread, the system hangs, no service responds
anymore, but the server still replies to ICMP and does not panic. I waited for
about 4 hours to see if the operations would proceed but they didn't.

--
Emile `iMil' Heitor .°. 
_
  | http://imil.net| ASCII ribbon campaign ( )
  | http://www.NetBSD.org  |  - against HTML email  X
  | http://gcu.info|  & vCards / \

Re: [RAIDframe] system hangs while manipulating large files

[Michael van Elst]

i...@home.imil.net ("Emile `iMil' Heitor") writes:

>/dev/dk0 on /export type ffs (asynchronous, log, noatime, NFS exported, local)

asynchronous + log should be impossible.

Did you find out wether the system is really hanging or just very slow?
In particular, is there any disk activity and does the system respond
on the network?

Re: [RAIDframe] system hangs while manipulating large files

[Thor Lancelot Simon]

On Fri, Jan 03, 2014 at 07:38:21PM +0100, Emile `iMil' Heitor wrote:
> 
> Could one of those mount flags cause any harm with a RAIDframe array ?
> 
> /dev/dk0 on /export type ffs (asynchronous, log, noatime, NFS exported, local)

I am surprised you can combine async and log but I would strongly
recommend against it.  I also suspect, since async removes many of
the constraints which would cause page cleaning, that using async
at all may be what's getting your system wedged.

Re: [RAIDframe] system hangs while manipulating large files

[Emile `iMil' Heitor]

Could one of those mount flags cause any harm with a RAIDframe array ?

/dev/dk0 on /export type ffs (asynchronous, log, noatime, NFS exported, 
local)


well well, it seems like removing the `async' flag solved the issue, I've just
ran multiple tests that caused instant fail yesterday which passed right away
without asynchronous I/O... some kind of memleak maybe ?

--
Emile `iMil' Heitor .°. 
_
  | http://imil.net| ASCII ribbon campaign ( )
  | http://www.NetBSD.org  |  - against HTML email  X
  | http://gcu.info|  & vCards / \

Re: [RAIDframe] system hangs while manipulating large files

[Emile `iMil' Heitor]

This value can be set to zero, however up to a factor of three
in throughput will be lost over the performance obtained at a 5%
threshold.


hm, I missed that point. Will raise up.


Well, I have applied both recommendations, and yet the machine just had the same
behaviour while rsync'ing data to the newly created RAID5. Again, the system
was kind of "up" but no command could be typed and services did not respond
anymore. Everything was frozen.

Could one of those mount flags cause any harm with a RAIDframe array ?

/dev/dk0 on /export type ffs (asynchronous, log, noatime, NFS exported, local)

Just to be clear, the trivial dd(1) test as of the rsync took place locally
on the server, no network transfers yet.

--
Emile `iMil' Heitor .°. 
_
  | http://imil.net| ASCII ribbon campaign ( )
  | http://www.NetBSD.org  |  - against HTML email  X
  | http://gcu.info|  & vCards / \