DSL Router code upgrade.

Tue Dec 19 17:10:22 PST 2006 — DSL Router code upgrade. At 12:01 AM Thursday, we will be pushing code upgrades to the majority of our DSL aggregation routers. Expected downtime for all customers is less than 5 minutes as the routers reload onto new code. -Jared

Update Thu Dec 21 01:13:09 PST 2006 — DSL Router code upgrade completed. We have completed the code upgrade to our DSL aggregation routers with no difficulties. Total downtime for any affected customers was less than 5 minutes. -Jared and Zeke

On Thursday December 21st Sonic.net will…

Mon Dec 18 13:07:54 PST 2006 — On Thursday December 21st Sonic.net will begin restricting traffic over port 25 on our dial-up services. We are making this change as part of our ongoing efforts to to improve our email performance and to fight the spread of spam. If you are curious about what this change means and why it is a good thing, see www.sonic.net/support/faq/advanced/port_25.shtml for details.

We believe that few if any of our customers will notice this change- except that everyone will ultimately benefit, because of the reduction in the spread of spam. — Sonic.net tech support

Sonic.net’s offices and technical support…

Wed Dec 13 19:19:12 PST 2006 — Sonic.net’s offices and technical support will be closed from 11:30am until 4:30pm PST on Thursday the 14th of December for our year-end staff party. If you should need assistance, please file a support ticket with support@sonic.net via email, or leave a voicemail on our support line at 707-547-3400. Dedicated access and colo customers can always reach us 24×7 on the operations hotline, and someone will be called away from the event to assist if required. -Dane

Broadlink Wally Moore tower problems.

Sat Dec 9 15:21:47 PST 2006 — Broadlink Wally Moore tower problems. There are fluctuating power issues with utility power at the Wally Moore site. Unfortunately this means that there could be further problems. All Broadlink Wally Moore connections should be working at this time. -Ron with Broadlink, and Support

Sacramento DSL issues.

Wed Dec 6 09:47:42 PST 2006 — Sacramento DSL issues. We have identified an issue inside of our transport vendor’s network that is causing intermittent connectivity for all DSL customers in the Sacramento area. Presently, the issue is manifesting itself as short periods (~5-10 seconds each) of zero connectivity occurring every 15 minutes. Our vendor has identified the issue, but has not yet provided us with an estimated repair time frame. We’ll keep this entry updated with information as we learn more. -Nathan

Update Wed Dec 6 10:27:48 PST 2006 — A temporary fix has been put in place to resolve this issue. A permanent solution is in the works. -Nathan

Update Fri Dec 8 00:56:17 PST 2006 — Sacramento DSL issues redux. A recurrence of issues in our transport vendor’s network caused a loss of connectivity to our Sacramento DSL customers from approximately 12:05 to 12:45 Friday morning. We continue to work with the provider for a permanent fix to this issue. -Jared and Nathan

Core router maintenance.

Fri Dec 1 14:20:05 PST 2006 — Core router maintenance. Tomorrow morning at 12:01AM we will be performing maintenance on one of our core routers located in San Francisco. This operation may interrupt some amount of Internet traffic for a few minutes. -Nathan and Zeke

Some AT&T DSL customers of all ISPs…

Tue Nov 21 11:10:30 PST 2006 — Some AT&T DSL customers of all ISPs (including Sonic.net) are currently experiencing slow performance that appears to be related to a network congestion issue in AT&T’s ATM network in California.

The congestion appears to have begun this morning around 3am, apparently when a number of customer routes were moved to a pair of circuits which are over capacity. This has resulted in packet loss and high latency for end-users whose links are being routed over the congested links. Estimated time to repair is one to three hours. – Eli, Nathan **I have now mailed this out with incorrect formatting twice. Please accept my apologies for the noise, though I might add that my co-workers are tremendously amused by this. – Eli

Santa Rosa DSL Migration.

Mon Nov 13 18:13:21 PST 2006 — Santa Rosa DSL Migration. At 12:01am on Wednesday, November 15th we will be migrating all of our DSL subscribers terminating in Santa Rosa to a new aggregation circuit landing in San Francisco. The new OC-12 is four times faster than the existing OC-3. In addition, this change brings our subscribers closer to AT&T’s network core. Downtime should be under 15 minutes, with most folks noticing only a 5 minute outage. -Nathan, Kelsey, Jared and Zeke

Update Wed Nov 15 02:21:27 PST 2006 — The DSL migration was completed successfully. -Nathan, Kelsey, Jared and Zeke

SpamAssassin Upgrades and Changes.

Fri Nov 10 14:24:55 PST 2006 — SpamAssassin Upgrades and Changes. Over the last couple of weeks we began testing a new SpamAssassin module that uses OCR (Optical Character Recognition) to scan attached images for known spam phrases resulting in a silent deployment last week. So far it, along with another 3rd party rule and score set, has proven to be very affective at catching most of the spam that was leaking through with an insignificant number of false positive matches. Also, in response to a user’s request, the ‘Drop Blacklisted Messages’ feature now also will drop messages matching Subject and ‘User in Blacklist To’ lists. -Kelsey

Mail Service Interruption.

Wed Nov 8 03:18:10 PST 2006 — Mail Service Interruption. For one hour and thirty four minutes this morning inbound mail delivery and retrieval was unavailable do to a failure on one of our four backend message stores. The exact cause of the failure is still unknown and now that service is restored we’re continuing to investigate the situation. This is the first time we’ve ever experienced a failure like this in the 10+ years that we have been using Network Appliance Filers. At this time all mail queued in our systems is delivering to customers mail spools. No mail was lost or not accepted for delivery for any reason related to this event by our MX servers during this outage. -Kelsey