Tue Mar 4 10:05:38 PST 2003 — Former Inreach DSL customers in the Stockton (LATA 9) area will be moved to the Sonic network during a maintenance window Thursday, 3/6/03, between 5 and 7 AM. The actual outage should be much shorter than that. -John and Nathan

Sun Mar 2 09:34:46 PST 2003 — Border router failure. A VIP in one of our core Cisco 7500 routers failed this morning causing routing instability until it was removed. This VIP was been used to provide redundancy in our core and is not needed for normal operation. We will be investigating why the the failure of one of the router’s redundant interfaces interfered with it’s operation; it should have gone relatively unnoticed. While the router was having difficulty, reachability to or from our network may have been spotty and some of our prefixes were dampened by remote networks. At this time, all of the router’s services have been fully restored and our network has returned to normal. -Kelsey and Nathan

Fri Feb 28 19:09:27 PST 2003 — Huge increase in website hosting bandwidth quota size, plus reduction in fee for additional hosting bandwidth usage.

I am excited to announce that retroactively effective from February 1st, we are increasing the amount of data transfer included with multihomed hosting from one gigabyte per month to twenty gigabytes per month. We have also reduced the quota overage fee for regular and multihomed accounts from $15.00 per gigabyte transfered to $4.95 per gigabyte.

This higher quota for multihomes and lower bandwidth cost for all hosted websites will be reflected on customer bandwidth invoices which will be posted in the first week of March for usage in February.

Discussions regarding hosting are underway at:

news://news.sonic.net/sonic.help.www

-Dane, Matt, John, Nathan, Kelsey, Eli, etc.

Fri Feb 28 16:41:12 PST 2003 — Tech Support hours extended. We’re pleased to announce extended Support hours! Sonic Support is now available from 8:00am-10:00pm on Weekdays, and 9:00am-10:00pm on Weekends. Support is closed every Friday from 1:00pm-2:00pm for our weekly meeting. -Eli, Sonic Support Crew

Thu Feb 27 14:33:33 PST 2003 — Various Updates and Fixes: We fixed a chronic issue that some MySQL users were experiencing yesterday regarding system resource consumption. We also brought a new dedicated mail server online to handle backup MX for customers as part of our preparation for transition to our new mail architecture. The new server also resolves issues we had in properly support ETRN on our mail cluster, it’ll now behave as expected. All affected customers received an email detailing the changes. We’ve also brought a new transit provider up at Equinix to reduce contention on our T3’s to uu.net and Cable and Wireless. -Kelsey, Nathan, Kevan and John.

Sat Feb 22 19:01:40 PST 2003 — Load balancer woes. We had quite a few intermittent failures of the load balancer link today, upon which we worked throughout the day to try to isolate the cause. We may have isolated the problem to a faulty cable, which has been replaced. We are monitoring to see if this change has fixed the problem. -Scott and Nathan

Update Sat Feb 22 20:30:40 PST 2003: Nope, that didn’t solve the problem. We have moved the load balancer off of the switch it was on and onto a core switch port, which is made by a different vendor. Hopefully this will at least eliminate possible cross-vendor incompatibility… -Scott and Nathan

Fri Feb 21 10:32:23 PST 2003 — More failures of the load balanced services this morning. We tweaked some knobs and changed some ports in our continuing effort to isolate the problem. -Kelsey and John

Mon Feb 17 17:27:17 PST 2003 — Periodic Failures of load balanced services. Rebooting of the AD3 and Summit48 haven’t resolved the periodic failures. We are going to work on them again tonight to see if we can’t restore stability until we have migrated off of this equipment. These failures will be most noticeable as failed connections to our mail servers and have been very transient. -Kelsey and Nathan

Sat Feb 15 18:48:05 PST 2003 — Night Operations: Monday morning, starting at 12:00AM, we are moving the T3 that connects our Santa Rosa HQ data center with our San Francisco pop into a different router in Santa Rosa. This will cause a momentary disruption in services while our OSPF and BGP routing tables converge. We are also going to reboot two other pieces of equipment: the last Extreme Networks switch in production and our Alteon AD3. These two have been having some issues cooperating and we hope that rebooting them will resolve the issues. These are the last remnants of our old network architecture and their replacements have already been tested and are waiting to be deployed as part of our mail server upgrades. While the two switches are rebooting, many of our services will be briefly unavailable. The replacements are, for the curious, two Cisco 3550’s, four Cisco 2924’s, two Alteon AD3’s and two Dell Powerconnect 5224’s. The new cluster of switches enables us to provide true redundancy to our entire network from end to end. -Kelsey and Nathan

Update: All planned maintenance was completed without a hitch.

Fri Feb 14 17:57:24 PST 2003 — The Screen Savers host Leo Laporte interviewed Sonic.net Co-founder, CEO and President Dane Jasper yesterday. For those of you who missed it, we have placed Windows Media Player files of the interview online at home.sonic.net/whatsnew/techtv/screensavers.shtml -John F