The gateway was not handing out leases and the wifi was down tonight.
The wifi was down because someone had switched off the wifi management device.
The main PoE switch for the public network seemed to be having issues.
I would plug in my laptop directly and the gateway and it wouldn't
give me a lease. Then I would plug into the gateway directly with my
laptop and it gave me a lease.
I moved all ethernet cables with their link-lights on to the private
switch and disconnected all cables there that were previously in the
private switch.
The public network is now working again but we need to figure out
what's up with that switch.
--
marc/juul
I called LMI and unfortunately they don't have gigabit in our area.
They can provide coverage over Sonic's fiber, but the fiber in our area is AT&T.
Sonic will re-sell AT&T fiber but LMI will not.
This means that it would cost $300 per month for fiber via Sonic and
it would be AT&T fiber.
What do folks think?
--
marc/juul
The gateway has been replaced with a desktop computer sitting on its
side in the rack on top of the multi-hd storage server. Note that this
is super janky since the two PCI network cards can't be mounted
properly in the somewhat-smaller-than-standard case.
We added two network cards and were planning to use the built-in card
as well, but currently the built-in card is not used and the private
network is down. There are two reasons:
* The switch used for the private network had one of its fans die and
I turned it off until we can replace the fan so it won't overheat
* The built-in network card appears to need firmware which I'm not
going to bother with until the fan on the private network switch is
replaced
Also, we might get a new server rather than using this old desktop machine.
The new computer is using the old harddrive. I changed the following
files due to the different names of the network interfaces:
/etc/network/interfaces
/etc/default/isc-dhcp-server
/etc/iptables/rules.v4
/etc/iptables/rules.v6
You can find backups of the files from before my changes at the same
paths with .old added (for the parent dir for /etc/iptables)
--
marc/juul
The internet is down.
Both the gateway is down and the wifi access points are not showing
up. I have not yet troubleshooted why the access points are down but
that may be simply because the gateway is gone.
The gateway overheated and may be dead. It had no CPU fan and no case
fans. It only had a small fan on the northbridge and that fan wasn't
working.
We have not been able to get it to boot even after letting it cool,
resetting the CMOS and installing an oversized fan.
We have a couple of more things to try but it looks like it's dead. I
will try to swap in a different computer.
--
marc/juul
We could have gigabit internet at Omni!
LMI sent this almost 2 months ago and I'm not sure how to reply
because the account is still in Marc Juul's name.
Is there a way to change the account to give multiple people access to it?
I just passed a Sonic truck on Shattuck on my way over here today.
They're putting optical splitters to peoples' houses less than a mile
up the road. It's time for us to get some of this. :)
---------- Forwarded message ---------
From: LMi Info <info(a)lmi.net>
Date: Thu, Mar 7, 2019 at 1:13 PM
Subject: [omni-accounts] Upgrade to blazing fast Phiber Optic Internet
To: <lmi-bulletin(a)lists.lmi.net>
Hello,
We're reaching out to let you know that our fiber optic internet service
is available, or will be available soon, at your address. Our Phiber
service is already available in parts of Albany, Kensington, El Cerrito,
and North Berkeley. The rest of Berkeley as well as North Oakland (north
of HWY580 and HWY24) is scheduled for delivery in the coming months!
There have been some delays but those challenges have been addressed.
As our customer, you are eligible for a service change to our new
blazing fast Phiber service at your home or business. You can find out
more about the service here:
https://www.lmi.net/phiber
If you would like to place a Phiber order, please respond to this email
and we'll get you up and running on our new service as soon as possible.
If you have any questions, please don't hesitate to contact our sales
department at 510-843-6389 ext. 1. Please note, Phiber may not be
available at your location.
Thank you for being an LMi.net member. We sincerely appreciate it.
--
LMi.net
510-843-6389
1700 Martin Luther King Jr. Way
Berkeley CA, 94709
I don't understand this message. It looks like a phish, but the site just redirects you to gandi's actual site. Maybe it's browser-specific or something.
* About to connect() to barraplaza.com port 80 (#0)
* Trying 31.193.131.100...
* Connected to barraplaza.com (31.193.131.100) port 80 (#0)
> GET /sudoroom.org HTTP/1.1
> User-Agent: curl/7.29.0
> Host: barraplaza.com
> Accept: */*
>
< HTTP/1.1 200 OK
< Content-Type: text/html; charset=UTF-8
< Server: Microsoft-IIS/8.5
< X-Powered-By: PHP/5.6.31
< Date: Fri, 24 May 2019 08:06:21 GMT
< Content-Length: 65
<
<script>
window.location="https://gandi.net";
</script>
* Connection #0 to host barraplaza.com left intact
‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐
On Thursday, May 23, 2019 9:53 PM, support(a)mail.gandi.net <mo7(a)basmaji.fr> wrote:
> Bonjour,
>
> Nous n'avons pas reçu le paiement sur votre dernière renouvellement du domaine sudoroom.org.
> Le solde reste endetté de 5,00 €.
>
> Pour éviter le problème, on vous invite à remplir manuellement le formulaire de renouvellement de
> vos services suivant les instructions et les étapes sur le lien ci-dessous:
>
> [Accédez à votre formulaire de paiement.](http://barraplaza.com/sudoroom.org)
>
> IMPORTANT : En cas de non règlement sous 48 Heure, votre compte pourrait être définitivement effacé.
>
> N'hésitez pas à contacter notre service client en cas de problème ou pour toute autre
> question : http://www.gandi.net/faq/contact_support
>
> Nous vous remercions de votre confiance.
>
> Cordialement.
>
> --
>
> GANDI - http://www.gandi.net/
The culprit was the php5-fpm process. Restarting it fixed it.
Error output from apache:
[Mon May 20 15:36:53.945158 2019] [proxy_fcgi:error] [pid 20480:tid
139913676658432] [client 50.57.61.7:62769] AH01067: Failed to read
FastCGI header
[Mon May 20 15:36:53.945213 2019] [proxy_fcgi:error] [pid 20480:tid
139913676658432] (104)Connection reset by peer: [client
50.57.61.7:62769] AH01075: Error dispatching request to :
Looks like someone set up some kind of status checker that locally
hits blah.sudoroom.org every second?
2019-05-20 15:39:45 127.0.0.1 blah.sudoroom.org "GET
/server-status?auto HTTP/1.1" 200
Was it supposed to inform us that sudoroom.org was down?
--
marc/juul