perfSONAR User Q&A and Other Discussion

[Tao Zhang <>]

Hi team,

I have been struggling with a weird OWPING problem related to two perfSONAR nodes:

http://lat-pe.upei.ca/toolkit/ (UPEI)  and  http://lat-vctr3pfs1.canarie.ca/toolkit/ (Victoria)

 

Both nodes are running the latest perfSONAR toolkit.

 

From UPEI node: (Bad)

owping lat-vctr3pfs1.canarie.ca

Approximately 13.5 seconds until results available

owping: FILE=protocol.c, LINE=1270, _OWPReadTestRequest: Unable to read from socket.

owping: FILE=owping.c, LINE=1838, Unable to fetch data for sid(cdbd209ede567ff0087b7aca7ccba298)

 

--- owping statistics from [lat-vctr3pfs1.canarie.ca]:8960 to [lat-pe.upei.ca]:9227 ---

SID:        8995f7fade567ff01281a1d6f76b3ec7

first:       2018-03-16T14:40:01.474

last:        2018-03-16T14:40:10.637

100 sent, 0 lost (0.000%), 0 duplicates

one-way delay min/median/max = 38.2/38.3/38.6 ms, (err=0.257 ms)

one-way jitter = 0.1 ms (P95-P50)

Hops = 8 (consistently)

no reordering

 

From Victoria node: (Good)

owping lat-pe.upei.ca

Approximately 13.5 seconds until results available

 

--- owping statistics from [lat-vctr3pfs1.canarie.ca]:8910 to [lat-pe.upei.ca]:8855 ---

SID:        8995f7fade568305e42eee5d97950e8c

first:       2018-03-16T10:53:11.344

last:        2018-03-16T10:53:21.488

100 sent, 0 lost (0.000%), 0 duplicates

one-way delay min/median/max = 38.1/38.2/38.5 ms, (err=0.251 ms)

one-way jitter = 0.1 ms (P95-P50)

Hops = 8 (consistently)

no reordering

 

--- owping statistics from [lat-pe.upei.ca]:8956 to [lat-vctr3pfs1.canarie.ca]:9880 ---

SID:        cdbd209ede568305ee3fd1eb770ecabd

first:       2018-03-16T10:53:11.337

last:        2018-03-16T10:53:20.639

100 sent, 0 lost (0.000%), 0 duplicates

one-way delay min/median/max = 38.2/38.3/38.7 ms, (err=0.251 ms)

one-way jitter = 0 ms (P95-P50)

Hops = 8 (consistently)

no reordering

 

The weird thing is that both nodes are working well with other nodes. The problem only happens in one direction when doing the OWPING test from UPEI node.

 

I did some curl testing from the UPEI node, apparently, it cannot access the archive of Victoria node. But it doesn’t have any problem accessing other nodes’ archives.

 

curl -v http://lat-vctr3pfs1.canarie.ca/esmond/perfsonar/archive/

* About to connect() to lat-vctr3pfs1.canarie.ca port 80 (#0)

*   Trying 205.189.32.158...

* Connected to lat-vctr3pfs1.canarie.ca (205.189.32.158) port 80 (#0)

> GET /esmond/perfsonar/archive/ HTTP/1.1

> User-Agent: curl/7.29.0

> Host: lat-vctr3pfs1.canarie.ca

> Accept: */*

> 

* Recv failure: Connection reset by peer

* Closing connection 0

curl: (56) Recv failure: Connection reset by peer

 

Any ideas?

 

 

Thanks

 

Tao