[Gluster-users] Issue in reading data using fio + gfapi

[Prashanth Kumar N]

Hi,
Digging some more, dropping Linux cache after every write and then doing read, makes my iostat show values signifying that reads are happening from disk.
This is quite strange as I have used direct=1 and also performance.strict-o-direct is ON. Doesn't this both option disable Linux page cache?
Any inputs on this would be useful.

-Prashanth

From: Prashanth Kumar N
Sent: 21 January 2016 14:36
To: 'gluster-users at gluster.org'
Subject: Issue in reading data using fio + gfapi

Hi,

I have been trying to use gfapi to read files written using fio  gfapi engine,  but don't see anything getting read. When I check disk activity using iostat or low level tools for my PCIe SSD, there is no activity.

Is the below sequence of fio correct way of using it? Am I missing something?

[root at vsan-blr4 prashanth]# fio --name=seq_write.1.0 --filename=seq_write.1.0 --rw=read --ioengine=gfapi --direct=1 --bs=4K --iodepth=1 --numjobs=8 --group_reporting --volume=pvol --brick=192.168.1.3 --time_based=40 --size=1G

fio: time_based requires a runtime/timeout setting
seq_write.1.0: (g=0): rw=read, bs=4K-4K/4K-4K/4K-4K, ioengine=gfapi, iodepth=1
...
fio-2.2.10
Starting 8 processes
Jobs: 8 (f=0): [R(8)] [inf% done] [0KB/0KB/0KB /s] [0/0/0 iops] [eta 1158050441d:07h:00m:13s]

[ Note: As soon as I start the fio job, I see a huge ETA number which seems fishy and fio exits in less than 5 seconds. ]


Thanks
Prashanth

HGST E-mail Confidentiality Notice & Disclaimer:
This e-mail and any files transmitted with it may contain confidential or legally privileged information of HGST and are intended solely for the use of the individual or entity to which they are addressed. If you are not the intended recipient, any disclosure, copying, distribution or any action taken or omitted to be taken in reliance on it, is prohibited.  If you have received this e-mail in error, please notify the sender immediately and delete the e-mail in its entirety from your system.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.gluster.org/pipermail/gluster-users/attachments/20160126/9037da7a/attachment.html>