cloyne / network Goto Github PK

Jalissa can't connect in W3I

Two Supermicro devices? Maybe some management interface for servers we have?

(They are visible even now when we moved the network to a new IP pool.)

I had to renumber the network again. All APs should be re-labeled.

Parisa reports:

Earlier tonight, it would allow me to connect but then would say "limited" and not work. It would not
disconnect me, but sometimes it would have 4 bars and still not work. Right now it is working and it has 4 bars. I don't know why it does that sometimes ha!

The list of rooms with reports of ethernet ports not working:

• C1E
• C2K
• E3F
• E2E
• E2D
• C3A

This ticket serves to document the process of replacing Cisco APs with UniFi APs.

We replaced all of them with UniFi AP PRO on February 1 2015. We kept the same channels as they were before.

But there are issues. Pings are not instead 10 ms more around 300 ms very often. Also things simply stop working for clients, despite them being connected to the AP.

I configured now that all APs have "medium" (21 dbm) transmission power.

I also configured RSSI, so that clients with bad signal will be kicked off. config.properties is now:

config.minrssi.0418d6206044.ng=25
config.minrssi.0418d62058df.ng=25
config.minrssi.0418d6205d20.ng=25
config.minrssi.0418d6205cfe.ng=25
config.minrssi.0418d62059c8.ng=25
config.minrssi.0418d6205a1f.ng=25
config.minrssi.0418d62059c7.ng=25
config.minrssi.0418d62058e4.ng=25
config.minrssi.0418d6205fe6.ng=25

So that we can decide which movies are popular and we should organize screenings.

• https://github.com/chilldude/motion
• https://github.com/TelescopeJS/Telescope

One broke. We should return it and get a new one.

Once ownCloud is working, the ability to attach personal hard drives in cloynes ownCloud to increase our personal storage space

We should install Farnsworth, developed at Kingman last year. It is a system for internal house's workflows. We could also add to it ideas we had, like a directory of all members (if it is not already implemented).

Was reported in the Thursday morning, we have to check if it is better now after restart.

Label all APs with their IPs (put a sticker).

Ren and Stimpy's printer is running out of black ink.

Here are some links

White Cat6 1000' riser cable $159.99 Amazon Prime
black cat5e riser cable 1000' $128.01 no amazon prime
100 cat5e modular plugs no amazon prime $3.78
200 cat6 modular plugs $24.99 amazon prime
100 multi colored boots $8.15
100 single random color modular plug boots $4.59 no amazon prime

The Ethernet cable connected to the computer in the kitchen does not work.

Low download with speedtest (1-3 Mbits).

Member can connect to the WiFi network, but no page loads.

I had to renumber the network again. All switches have to be moved to new IPs.

We got reports that WiFi is being disconnected in E3E, E3F.

There used to be an access point near Graham's office that used to also be an Ethernet hub but now there is only a female-female adapter and a hanging wire.

Cold restarting all the cisco routers via web-ui.

Confirming with people if problem is solved, then hopefully closing tickets.

Kitchen and especially kitchen computer have spotty Internet access. The idea is to move the only ethernet cable which goes currently to the AP in the great hall into Grahams office, put there one mode node, and then from there take multiple cables, one to the stage, one to AP, one to kitchen, where we would put another AP and then connect the kitchen computer to that AP.

Since last week. And the WiFi is pretty bad.

There is currently a 100mb switch in the C3 network box, and by the office in E1. Should we replace them with gigabit switches?

Currently we have a firmware from October 2014 April 2015.

• FamilyRoom
• W3J
• E2J
• W2C
• W3A
• Hackerspace
• E3StudyRoom
• GreatHall
• E2D
• W2H
• C2P
• C2C
• W1E
• C3K

The ability to e-mail the printer at [email protected]/org to be able to print so we don't have to print through that randomly generated string thing that no one remembers

Member reported that they have issues with Internet in W2H room. It is strange because ping is lossy. So it is not down, but it is lossy. Around 20% packet loss. But enough that I cannot connect to the AP using SSH.

The internet connection in my room has not been working very well lately. On my laptop it says that I'm connected but I wasn't even able to send you an email. I had to use my phone. Could you check it out? My room is W2D and if I'm not here it's fine, you can go in.

Since today or maybe since I added 3 new nodes on Sunday morning, the network does not behave like network anymore and for example I cannot ping from one location all WiFi points. But they work. And if I go to them I can ping them directly. It seems like we are not in one common network anymore. No idea why not.

Could it be because I moved some APs around and now switches are confused? Did this happen in the past?

There are many loose / unplugged cables.

So that people from Yahoo can send e-mails to the mailing list. See: http://www.sympa.org/manual/dmarc

I like to ssh, especially since I have such a weak little computer, and that's hard to do with a dynamic IP

I've yet to test WiFi in there since the new firware flashes on the unifis but that might also still be an issue; I'll update. Either way, Ethernet in there would be very nice.

Switches should be configured that DHCP traffic from non-router cannot go through. We should check all switches and see if this was configured. We should check if we can configure this per each port (so no non-router DHCP traffic can spread even around the same switch). Also, now that we moved IP range of the router, do we have to change anything on switches? Or is the rule based on the MAC address?

I made iperf measurements between server2 and all APs on ethernet. On server2 I ran:

iperf -s

And then on each AP I ran:

iperf -c 10.20.32.10

Results:

Server listening on TCP port 5001
TCP window size: 85.3 KByte (default)
------------------------------------------------------------
[ ID] Interval       Transfer     Bandwidth
[  4] local 10.20.32.10 port 5001 connected with 10.20.32.40 port 51919
[  4]  0.0-10.0 sec   112 MBytes  94.1 Mbits/sec
[  5] local 10.20.32.10 port 5001 connected with 10.20.32.41 port 37307
[  5]  0.0-10.0 sec   113 MBytes  94.1 Mbits/sec
[  4] local 10.20.32.10 port 5001 connected with 10.20.32.42 port 35974
[  4]  0.0-10.0 sec   161 MBytes   135 Mbits/sec
[  5] local 10.20.32.10 port 5001 connected with 10.20.32.42 port 35975
[  5]  0.0-10.0 sec   155 MBytes   130 Mbits/sec
[  4] local 10.20.32.10 port 5001 connected with 10.20.32.43 port 37897
[  4]  0.0-10.0 sec   153 MBytes   128 Mbits/sec
[  5] local 10.20.32.10 port 5001 connected with 10.20.32.44 port 45680
[  5]  0.0-10.0 sec   112 MBytes  94.1 Mbits/sec
[  4] local 10.20.32.10 port 5001 connected with 10.20.32.45 port 37447
[  4]  0.0-10.0 sec   199 MBytes   167 Mbits/sec
[  5] local 10.20.32.10 port 5001 connected with 10.20.32.46 port 42214
[  5]  0.0-10.0 sec   208 MBytes   175 Mbits/sec
[  4] local 10.20.32.10 port 5001 connected with 10.20.32.47 port 40292
[  4]  0.0-10.0 sec   112 MBytes  93.9 Mbits/sec
[  5] local 10.20.32.10 port 5001 connected with 10.20.32.48 port 53549
[  5]  0.0-10.0 sec   167 MBytes   140 Mbits/sec
[  4] local 10.20.32.10 port 5001 connected with 10.20.32.49 port 42186
[  4]  0.0-10.0 sec   156 MBytes   130 Mbits/sec
[  5] local 10.20.32.10 port 5001 connected with 10.20.32.50 port 55565
[  5]  0.0-10.0 sec   112 MBytes  94.0 Mbits/sec
[  4] local 10.20.32.10 port 5001 connected with 10.20.32.51 port 60262
[  4]  0.0-10.0 sec   150 MBytes   125 Mbits/sec
[  5] local 10.20.32.10 port 5001 connected with 10.20.32.52 port 46483
[  5]  0.0-10.0 sec   111 MBytes  93.0 Mbits/sec
[  4] local 10.20.32.10 port 5001 connected with 10.20.32.53 port 47252
[  4]  0.0-10.0 sec   134 MBytes   112 Mbits/sec

On APs TCP window size was 21.0 KByte (default).

It seems our network is nice 100 Mbit network. Some connections are more, but not really gigabit. :-) But I think it is good.

@sreezon101, can you look into this?