Hi,

since about 20 minutes (from 13:40:00 UTC), SBG1 object storage
exhibits very high latencies on simple GET requests. Initially, it was
frequent and high peak latencies. Meanwhile, it looks like 100% of the
requests suffer from the very high latencies.

I've attached a graph from our monitoring that shows the dramatically
deteriorated response times.

And also apache bench shows it:

--

ab -n 50 -c 5
http://storage.sbg1.cloud.ovh.net/v1/AUTH_1242XXXXXXXXXXXXXXXXX/test-videos/squid_test.txt
This is ApacheBench, Version 2.3 <$Revision: 655654 $>
Copyright 1996 Adam Twiss, Zeus Technology Ltd, http://www.zeustech.net/
Licensed to The Apache Software Foundation, http://www.apache.org/

Benchmarking storage.sbg1.cloud.ovh.net (be patient).....done


Server Software:
Server Hostname:        storage.sbg1.cloud.ovh.net
Server Port:            80

Document Path:
/v1/AUTH_1242dXXXXXXXXXXXXXXXXXXXXXXXXXXX/test-videos/squid_test.txt
Document Length:        20 bytes

Concurrency Level:      5
Time taken for tests:   33.955 seconds
Complete requests:      50
Failed requests:        0
Write errors:           0
Total transferred:      22850 bytes
HTML transferred:       1000 bytes
Requests per second:    1.47 [#/sec] (mean)
Time per request:       3395.476 [ms] (mean)
Time per request:       679.095 [ms] (mean, across all concurrent requests)
Transfer rate:          0.66 [Kbytes/sec] received

Connection Times (ms)
               min  mean[+/-sd] median   max
Connect:        0    0   0.0      0       0
Processing:  3032 3089  82.5   3058    3390
Waiting:     3032 3089  82.5   3058    3390
Total:       3032 3090  82.5   3058    3390

Percentage of the requests served within a certain time (ms)
   50%   3058
   66%   3071
   75%   3084
   80%   3089
   90%   3237
   95%   3317
   98%   3390
   99%   3390
  100%   3390 (longest request)

--

 From what I see, there is no incident open about this issue yet. Are
you aware of the problem?

Best regards,
Michael