On 30.04.15 19:51, Gene Heskett wrote:
> But I have a PITA in the machine that runs my lathe, its recently turned
> into a crashomatic, with uptimes of about an hour!
>
> And since it is an nfs mount, when it crashes, it locks up the rest of
> the machines that are mounting it. Thats the PITA problem. And IMO a
> damned bug in nsf4.
Gene, if the problem is not just loss of access (unavoidable once the
NFS exporter has crashed), but the lock-up you describe, then changing
the NFS mount from "hard" to "soft" should fix that.
("man nfs" says data integrity may suffer if connection is not over
TCP-IP, but I've never noticed, admittedly with now 30-year old NFS,
under Solaris)
That manpage does, though, say: "Using the intr option is preferred to
using the soft option because it is significantly less likely to result
in data corruption." But then it goes on to say it isn't much use after
kernel 2.6.25. Looks like the developers don't use NFS much.
In the old days, I used soft mounts to allow a server farm to come up
despite NFS cross-mounts. With hard mounts, and A needing B, and B
needing A, they could never come up. Using a soft mount on one side let
them boot, at the cost of a manual NFS mount a little later. (Soft
mounts on both sides could necessitate two manual NFS mounts)
Erik
--
There's no obfuscated Perl contest because it's pointless.
-Jeff Polk
------------------------------------------------------------------------------
One dashboard for servers and applications across Physical-Virtual-Cloud
Widest out-of-the-box monitoring support with 50+ applications
Performance metrics, stats and reports that give you Actionable Insights
Deep dive visibility with transaction tracing using APM Insight.
http://ad.doubleclick.net/ddm/clk/290420510;117567292;y
_______________________________________________
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users
