Node name: Division 34
Status: Active
Live Date: February 9, 2004
Contact: RussellSenior
Website: http://www.hotpepper.com
Hot Pepper Studios
3377 SE Division St # 104
Portland, OR 97202
(503) 230-2058
Map: https://cwnmyr.personaltelco.net/nodes/HotPepper
Equipment
- Alix.2d3 (PTP's)
LinkSys WAP-11
- Apple Airport Base Station (Graphite)
- 150+ feet of ethernet cable
Software
Network Configuration
- ESSID: www.personaltelco.net
- Channel:
- BSSID: 00:0C:41:18:53:B5
- Hostname: lester
- Internet IP: 24.21.69.222 (configured by DHCP)
- PTPnet Network: 10.11.14.1/24
Maintenance and System Log
2009-07-30: Bad throughput on the node indicated network problems. Eventually JasonMcArthur and I solved it by setting eth0 (the WAN interface) to MTU of 576 (derived from google suggestions). That seems to fix the problems. Was seeing it on the lester-the-nucab prior to replacement also. Bad ISP! The MTU fix is kludged into /etc/init.d/network, since the supposed option in /etc/config/network ("option mtu <value>" didn't seem to override dhcp). If we find the "proper" fix, we should correct that. --RussellSenior
2009-07-30: MarinoDuregon stopped at location parking lot, associated to the base station, ssh'd into the AlixCab. From there the Cable Modem was reset through http management interface. Once the AlixCab had connectivity new OpenVPN keys were dropped into place. lester was rebooted and was back online. -- MarinoDuregon
2009-07-29: RussellSenior and MarinoDuregon replaced the NuCab and Linksys 5-port hub with an AlixCab. Wireless association through Airport base station working but having trouble with internet connectivity because Cable Modem that is remote and can't be manually reset does not recognize the new MAC address. Left for Weekly Meeting. -- MarinoDuregon
2008-06-23: Responded to an outage report. Cacti graphs imply the outage has been for several months, since the end of February-ish, but that's not certain. It looks as if it was triggered by an upstream dynamic IP address change, which rendered the upstream connection useless for some reason (shouldn't dhclient or equivalent handle that??). The outage when then compounded by /var filling up. The biggest culprit was /var/lib/logcheck/dead-letter, which apparently filled up with undeliverable logs of various services complaining about not having a network or the disk being full or something. Blew the logs away. Then also had a problem with nocatauth getting confused about the local network and guessing it was the openvpn tunnel. I think JasonMcArthur fixed that up and it was back in business. Signal was weak outside, pinged the node hosts about maybe re-evaluating the gear there. --RussellSenior
Notes
- External wireless coverage is poor. Building is shielding signal. Plans are to install external antenna on roof to irradiate the signal in the vicinity.
History
- By summer 2004, we expect this node to expand it's signal coverage throughout the Division 34 building and across the street to Pix Patisserie.
Fall 2004: coverage throughout Division 34 building. Still working out the connection to Pix. - RossOlson