[Openstack] Download file from swift extremely slow

dongjh at ahope.com.cn dongjh at ahope.com.cn
Thu Jan 5 03:18:07 UTC 2017


Hi Charles,

There's no any log file on the controller(swift proxy server), it's a Newton manual installation, just one controller.

[root at controller swift]# pwd
/var/log/swift
[root at controller swift]# ls
[root at controller swift]# cd ..
[root at controller log]# ls
anaconda       dmesg.old  maillog-20170101   secure-20170101
audit          firewalld  mariadb            spooler
boot.log       glance     messages           spooler-20170101
btmp           glusterfs  messages-20170101  swift
btmp-20170101  grubby     neutron            tallylog
chrony         horizon    nova               tuned
cinder         httpd      ppp                wtmp
cron           keystone   rabbitmq           yum.log
cron-20170101  lastlog    rhsm               yum.log-20170101
dmesg          maillog    secure


=========================================
董 建 华
地址:杭州滨江区南环路3766号新世纪办公楼
邮编:310053
手机:13857132818
总机:0571-28996000
传真:0571-28996001
热线:4006728686
网址:www.ahope.com.cn
Email:dongjh at ahope.com.cn
 
From: Charles Hsu
Date: 2017-01-05 11:13
To: dongjh at ahope.com.cn
CC: nasir.mahmood; Rick Jones; openstack
Subject: Re: [Openstack] Download file from swift extremely slow
Hi Dongjh,

I saw there is tran-id (`tx559e7d3d28aa49c995467-00586d03cf`) in your log, please try to find the swift log in your controller node and grep it likes `grep tx559e7d3d28aa49c995467-00586d03cf  /var/log/<swiftlog>` and paste output here.

It seems like you deploy this openstack env with Mirantis Fuel? How many controller nodes that you have on this test env?

Best,
Charles


2017-01-05 10:56 GMT+08:00 dongjh at ahope.com.cn <dongjh at ahope.com.cn>:
In the [filter:cache] section, configure the memcached location:
[filter:cache]
use = egg:swift#memcache
...
memcache_servers = controller:11211

memcache_server or memcached_server ?



=========================================
董 建 华
地址:杭州滨江区南环路3766号新世纪办公楼
邮编:310053
手机:13857132818
总机:0571-28996000
传真:0571-28996001
热线:4006728686
网址:www.ahope.com.cn
Email:dongjh at ahope.com.cn
 
From: Nasir Mahmood
Date: 2017-01-05 02:25
To: Rick Jones
CC: openstack
Subject: Re: [Openstack] Download file from swift extremely slow
Just a side note, can you please run the 'iostat -xnpz 1' on your swift nodes  and share results while downloading file , including network performance tests suggested earlier ... 

I would also suggest to share the logs on swift nodes would be much helpful on this by enabling debug in config.

On Jan 4, 2017 22:51, "Rick Jones" <rick.jones2 at hpe.com> wrote:
On 01/04/2017 12:59 AM, dongjh at ahope.com.cn wrote:
Hi experts

I finished the swift installation following the install
guide(http://docs.openstack.org/project-install-guide/object-storage/draft/get_started.html),
file upload  is very fast, but file download is extremely slow, why ?

[root at controller admin]#* time openstack object create container1 cirros-0.3.4-x86_64-disk.img*
+------------------------------+------------+----------------------------------+
| object                       | container  | etag                             |
+------------------------------+------------+----------------------------------+
| cirros-0.3.4-x86_64-disk.img | container1 | ee1eca47dc88f4879d8a229cc70a07c6 |
+------------------------------+------------+----------------------------------+

real    0m3.807s
user    0m2.127s
sys     0m0.161s


[root at controller /]#* time  openstack object save container1 cirros-0.3.4-x86_64-disk.img*
real    5m51.489s
user    5m48.172s
sys     0m2.094s

Are you able to run something like netperf or iperf between your client and the swift proxy?  For example:

netperf -H <proxy>    # get a feel for "to swift" basic network perf
netperf -H <proxy> -t TCP_MAERTS # get a feel for "from swift"

The idea there is to measure the network separate from the storage and swift processing, and go from there.

If there isn't much else happening on your setup at the time, you could also look at some snapshots of netstat -s on the proxy when you are downloading the object - look to see if there are many TCP retransmissions.  You can get something similar "directly" for the netperf tests with:

netperf -H <proxy> -- -o throughput,local_transport_retrans,remote_transport_retrans

netperf -H <proxy> -t TCP_MAERTS -- -o throughput,local_transport_retrans,remote_transport_retrans

rick jones

_______________________________________________
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to     : openstack at lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

_______________________________________________
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to     : openstack at lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack/attachments/20170105/6c823de0/attachment.html>


More information about the Openstack mailing list