Re: Question about IO workload via NBD
Yes. I enabled flush option as well. But then I figured out it is a
test setup issue. Thank you for the help!
I have another question.
I tried to reload the nbd server config file by sending SIGHUP to the
process. The client is about to setup the connection fine with the
server
sudo /usr/local/sbin/nbd-client xxxx -N test1 /dev/nbd1 -b 4096 -persist
Negotiation: ..size = 10240MB
bs=4096, sz=10737418240 bytes
But then when I tried to run FIO on that device, it complained that it
is a 0 size device.
/dev/nbd1: zero sized block device?
The nbd client syslog says
kernel: block nbd1: Connection timed out
kernel: block nbd1: shutting down sockets
kernel: print_req_error: 5 callbacks suppressed
kernel: print_req_error: I/O error, dev nbd1, sector 0
The nbd server syslog says
Connection dropped: Connection reset by peer
Is this a known issue? Is there a workaround for reloading the config
file without killing the existing connection?
I appreciate any help in advance.
Best wishes,
Taylor
On Sat, Sep 7, 2019 at 6:47 AM Wouter Verhelst <w@uter.be> wrote:
>
> On Tue, Sep 03, 2019 at 02:47:52PM -0700, Taylor Yang wrote:
> > Hi Richard,
> >
> > The command I use is like `fio --name=xxx --rw=randread --output=xxx
> > --direct=1 --ioengine=libaio --bs=16k --iodepth=4 --filename=/dev/nbd0
> > --size=1024Mb` and I setup both nbd client and nbd server on two
> > different hosts. The FIO job is executed on the nbd client side after
> > the connection between the nbd client and server being setup. I think
> > it is because nbd server uses the buffer cache but I also tried to
> > specify the `sync` option in the config file and it still behaves the
> > same way.
>
> Did you enable the flush and/or fua options?
>
> --
> To the thief who stole my anti-depressants: I hope you're happy
>
> -- seen somewhere on the Internet on a photo of a billboard
Reply to: