Hello Max!
Main server on 100/100 dedicated connection on servercenter.
International granted bw. min 20Mbit
I tested now:
lshw -C network
WARNING: you should run this program as super-user.
*-network
description: Ethernet interface
product: NetXtreme BCM5722 Gigabit Ethernet PCI Express
vendor: Broadcom Corporation
physical id: 0
bus info: pci@0000:0e:00.0
logical name: eth0
version: 00
serial: f4:ce:46:e6:63:d0
size: 100Mbit/s
capacity: 1Gbit/s
width: 64 bits
clock: 33MHz
capabilities: bus_master cap_list ethernet physical tp 10bt
10bt-fd 100bt 100bt-fd 1000bt 1000bt-fd autonegotiation
configuration: autonegotiation=on broadcast=yes driver=tg3
driverversion=3.121 duplex=full firmware=5722-v3.07 ip=195.56.44.167
latency=0 multicast=yes port=twisted pair speed=100Mbit/s
@server6:~$ ./speedtest-cli
Retrieving speedtest.net configuration...
Retrieving speedtest.net server list...
Testing from GTS Hungary Telecommunications Limited (195.56.44.167)...
Selecting best server based on ping...
Hosted by TvNetTel (Szentes) [43.37 km]: 11.285 ms
Testing download speed........................................
Download: 93.79 Mbit/s
Testing upload speed..................................................
Upload: 87.88 Mbit/s
The up/down link is good.
The log files send out nginx, not throtled.
The OM throtling bw only Red5 config, i not modifye this.
In red5 find bw settings, maximised 10 Mbit/client.
I try download log - on another server - from om server (100/100 dedicated).
$ wget http://195.56.44.167:8080/openmeetings/log/olds/jvm.stdout
--2014-02-26 13:20:24--
http://195.56.44.167:8080/openmeetings/log/olds/jvm.stdout
Connecting to 195.56.44.167:8080... connected.
HTTP request sent, awaiting response... 200 OK
Length: 24909958 (24M) [application/octet-stream]
Saving to: “jvm.stdout”
100%[======================================>] 24,909,958 11.2M/s in 2.1s
2014-02-26 13:20:26 (11.2 MB/s) - “jvm.stdout” saved [24909958/24909958]
The nginx based log bandwith is good. 11.2 MB/s ~ 100Mbps
This test consequent:
If red5 send unthrotled, the <2MByte app ~ downloadable max 20Mbps -
current settings dwl max (on 10 Mbps linit) 2 sec.
If client network 3 Mbps dwl bw, the 2 MByte app dwl time - on math base
- < 10 sec.
This results they are distant to the real app loading procedure 35~120
sec loading time.
Balazs
2014-02-26 12:29 keltezéssel, Maxim Solodovnik írta:
Hello Varga,
Your server is very slow (I was unable to download your log file due
to timeout)
You need faster inet connection on server for videoconferencing
On Feb 25, 2014 4:02 PM, "Varga Balázs" <[email protected]
<mailto:[email protected]>> wrote:
Hello Max!
I try OM server (now 3.1) loadings.
Main internet not boost, simple ADSL 5/0,5 Mbps, avg 3,7/0.37 Mbps
~ 0.4 MBps/0.04MBps.
The complete OM app (as moderator) download in non cached browser
~1.8 MB in many block. On main connection ~ 17 sec (onloan ~6sec)
The finaly point, app downloaded, initialized, rendering
completedi 59.48 sec!
The loader process try 1,2 good, try 3 is very slow! In try 3
flash/browser is freeze (grey layer on firefox, plugin not
response in chrome) temporary
If login as partizipant (registered/nonregistered eq.) after try3
point is set language panel and have plus many loading (language
specific content).
The partizipant loading time 1:40 !
How mutch the complete OM downolad size? Have optimize the app
size eg. images
The firefox listening only http query size, not measure ower
flash/rtmp download (theme.xml).
I don't understand, whie is many-many time the initialization. I
try this procedure over high sped cable net (60/20Mbps) and the
loading time not super speed into. What is a slowly "backend"
procedure?
The network speed is avg 1-5Mbps download in european users. The
partizipants not await above 30 sec on loading, leave the
conference and not apply this in futhure!
Howto reduce downloading sizes and speed up rendering, processing
ower flash process?
Balazs
