Node name: Lucky Lab
Live Date: 25 November, 2005
Sponsor: Lucky Lab Brewing Company
Contact: Gary Geist (owner)

915 SE Hawthorne Blvd.
Portland, OR 97214
(503) 236-3555
http://www.luckylab.com/html/story.html

Map: https://cwnmyr.personaltelco.net/nodes/LuckyLab

Equipment

• Two MR24 (installed around July 25, 2017)
• Ubiquiti ERX (installed around June 9, 2021)

Software Installed

• AlixCab

  • OpenWrt

  • NoCatAuth

  • OpenVpn

  • Olsrd
  • Snmpd

Network Configuration

• Host: fetch (erx gateway)
  • IPv4addr: 10.11.11.1/25
• Host: ap1 (meraki mr24)
  • IPv4: 10.11.11.2/25
• Host: ap2 (meraki mr24)
  • IPv4: 10.11.11.3/25

Installers / Organizers

• DonPark

• KeeganQuinn

• CalebPhillips

• RussellSenior

• TroyJaqua

• RobertPetersen

• Other People?

• Node is maintained by the NetworkOperationsTeam

Graphs

Maintenance and System Log

• 2021-06-09: Replaced Alix with Ubiquiti ERX (approximate date)
• 2017-07-25: Replaced Buffalo WZR600DHPs with Meraki MR24s (approximate date)

• 2013-09-06: Replaced the Netgear WGT634Us with Buffalo WZR600DHPs. One of the WGTs had been having what appeared to be thermal problems. Asked Gary about replacing them. Did that this evening. Now two dual-band routers, which each radio having three SSIDs. --RussellSenior

• 2013-02-13: Replaced the old WRT54Gs with Netgear WGT634Us that became surplus to some extent due to the recent router upgrades. These use atheros radios that can do multi-SSID, so I've set them up so that they can be distinguished (SSID's ending in /ap0 and /ap1, respectively). Also, I replaced the power splitter near ap0 (in the window under the surfboard) with something more robust (takes 48V down to 12V). Oh, also recrimped the end of the cat5e going to ap1 (over the door in the private room) which seemed to be flakey and intermittent. --RussellSenior

• 2013-01-25: Updated firmware to r35318 with linux kernel version 3.7.4 and IPv6 --RussellSenior

• 2010-04-09: After receiving reports recently of problems here, visited for the last weekly and observed weirdness. Decided to swap out the access points with WRTs from my basement, where we have a fair surplus. Installed the new OpenWrt release 10.03 using a 2.6 kernel and the new wireless driver, b43 with the mac80211 stack. Signal strength and stability seems much better. The northern WRT is in a window mounted on a special linksys mounting frame. Since it is inverted, I added some zip-ties to ensure it doesn't rattle out. --RussellSenior

• 2009-07-20: Stopped by to remember that we are using a static IP address here. I saw Gary and he described some networking problems they were having, possibly due to our dynamic IP address stomping on their static addresses. They use a Paradyn DSL modem from Integra and apparently its DHCP server isn't particularly smart about looking for existing usage of addresses before providing leases. Anyway, fixed up again. I told Gary which static IP we were using and suggested if they have continuing problems to call Integra and have them figure out where the leases are going, and maybe to trim down the lease range so as to not conflict with the static IPs. --RussellSenior

• 2009-07-18: RussellSenior and JorenLove remove the NuCab (Joren is hauling it to FreeGeek this evening), and plugged in our first deployed AlixCab. Was also able to remove an 8-port switch that isn't needed anymore. The Alix is bridging eth1 and eth2, so the two cat5 runs to the access points terminate there. --RussellSenior

• 2008-12-10: Got another report on ops that this node is down often. The several times I've found it broken recently have been a case of a missing default route. Down/up'ing the eth0 interface usually fixes it, but it looks like a problem maybe with the upstream DHCP server, so I have reconfigured the interface as static at a high-numbered address, hoping to avoid getting stomped on by the dsl modem's DHCP server. Let's see if that fixes it. The node is accessible on the openvpn/olsr network (i.e. from other nodes with working tunnels). We have cacti graphs for this node now too, at least so long as the network isn't broken. --RussellSenior

• 2008-11-06: Got a report from Gary that the node was down. He said they had power cycled stuff, but no workie. I arrived this evening to find that the nucab didn't have an address on eth0. Not sure why that would be. I ifdown/ifup'd eth0 and restarted nocatauth and it was back to working again. --RussellSenior

• 2008-06-10: Based on a report from StefanMintier, stopped to check on the node. It looks like there were some changes to our upstream network, which confused the nucab. I made a half-hearted attempt to fix it, which failed, then I rebooted the nucab, which succeeded. All is well now. --RussellSenior

• 2008-02-29: DLink over the south door in meeting room was freaking out. only allowing about 1k of data per tcp stream. powercycled the unit and all's well. --DonPark

• 2007-06-25: This node has had an outage for a couple weeks. I only heard about it by accident. I stopped by today to check on it. It turns out the cat5 run between the nucab and the dsl modem had been disconnected during some office work. Plugged back in and everything is golden. --RussellSenior

• 2007-01-24: Fixed default redirect after splash. Removed a bit of ancient info from the node page. Can't check back into svn because it is still down. -- CalebPhillips

• 2006-12-20: Found nucab with no default route. Apparently it comes up faster than the DSL modem. Bounced the interface, better now. --RussellSenior

• 2006-06-01: Found second (window) AP unpowered. Don't know why. No lights on PoE splitter or AP.

• 2006-04-28: Installed a power jumper from the PoE splitter to the second (window) AP. Lights on. Assuming function will follow. --RussellSenior

• 2006-04-19: Held WeeklyMeeting20060419 here. We finished the install of the second AP by re-crimping a poorly crimped end. -- CalebPhillips

  • ...or nearly finished the install. Sigh. Since our last visit, the power cord that ran from the POE splitter to the AP, which had been dangling, disappeared. I plugged in one from my stumbler rig to ensure it worked, but removed it before leaving. We need another power cord, basically two standard-ish 12V coaxial barrel connectors. Also, someone needs to forward ports (22, 5280, others?) from the DSL modem to the nucab.

• 2006-03-22: Held WeeklyMeeting20060322 here. We installed the second AP with PoE injector, but gave up when the cat5e run seemed to be bad (perhaps a kink?). Decision is to return some other time to fix it, perhaps when we have a better toolset. -- CalebPhillips

• 2006-03-01: Held a weekly meeting at the LuckyLab whose agenda included fixing some outstanding issues with this node. And, some of them were actually fixed! See WeeklyMeeting20060301 for details. -- CalebPhillips

• 2006-02-23: RussellSenior responded to an outage report. The power cord supplying the nucab (on top of the beer cooler in back) had been unplugged. Replugging fixed it.

• Filling in some missing documentation. Noticed that nodeluckylab.personaltelco.net doesn't resolve internally, which fudges the portal redirect. - CalebPhillips

Notes

• See also LuckyLab for directions and stuff.

• Good Beer!

• [CategoryNode][CategoryPersonalTelco][CategoryNodeBehindNat]