Month: January 1998

Fri Jan 16 10:16:51 PST 1998 — We had a problem with our authentication server this morning. Users were unable to authenticate for about 15 minutes. We apologize for any inconvience. -Brian

Tue Jan 13 17:03:34 PST 1998 — Pacific Bell’s reprovisioning of our hunt group to a directory hunt has fixed the capacity issues which have been a problem for the past few weeks. There are a couple remaining issues which we’re working on, but the bottom line is that phone lines and modems are now working and available! Thanks much for your patience as we’ve worked through this Pacific Bell issue. -Dane

Tue Jan 13 16:59:08 PST 1998 — Our news server has had a few crashes in the past couple days. With the kernel version we are running on this host, there is a problem which can lead to this type of failure in multiprocessor systems. A new version has been released (2.0.33) and are working on an upgrade. Because of these crashes, and the disk errors which result from them, we are running a bit backlogged on incoming news. This backlog should be cleared shortly. -Dane

Sat Jan 10 12:30:10 PST 1998 — On Sunday the 18th at 2am we will be upgrading the memory in our Network Appliance NFS filer. Downtime will be between 5 and 15 minutes. This downtime will affect many services. Local www, local ftp and mail will all be unavailable. Shell access will be available, but attempts to check mail will block until the filer is back online. We’re upgrading the filer all the way to it’s max capacity (4x it’s current configuration) so that we won’t have to do this again in the future. We can add and replace disks without taking the system offline, but memory upgrades require downtime, so we don’t want to do this again. -Dane

Sat Jan 10 12:26:21 PST 1998 — Pacific Bell completely reprogrammed the 522-1002 hunt group between 2am and about 6am last night. We hope that this will fix the problems which have been affecting that group. As it turns out, Pacific Bell did not contact NorTel support until yesterday, and there is a third type of group called directory hunting which must be used in groups with more than 100 channels. This has been implemented, and things appear to be working well. We will make an effort to simulate a high load condition before Monday night to test the reconfigured group all the way to its end. Thanks to Alma at PacBell and Scott here at Sonic for working the graveyard shift for us. -Dane

Sat Jan 10 19:46:29 PST 1998 — Our news server was offline briefly this afternoon after a crash. With the kernel version we are running on this host, there is a problem which can lead to this type of failure in multiprocessor systems. A new version has been released (2.0.33) and we’ll be upgrading shortly. -Dane

Fri Jan 9 15:57:58 PST 1998 — Pacific Bell believes that they have found the source of the trouble, and in order to fix it, they need to rebuild the entire hunt group. They will be doing this work starting at 2am on Saturday morning, and they expect to be finished by 5am. During this time, X2 and ISDN single-B access will be unavailable. (fingers crossed) -Dane

Thu Jan 8 17:14:07 PST 1998 — Pacific Bell found a programming error in the hunt group setup. Unfortunately, this was not the problem, and they are still looking for the source of the busy signals. I’ve spoken with Pacific Bell three times today, and I’ve also been working with the California Public Utilities commission. Because of ongoing issues with ISDN quality of service, a CPUC decision was reached last year – decision 97-03-021 allows us to collect damages from Pacific Bell for extended outages like this one. This is nice, as we’ve spent over $55,000 to equip and install the lines which should be serving you, and they’re not working!

We have filed an informal complaint with the CPUC, and they have contacted Pacific Bell on our behalf. I don’t know if this will help much, and we’re considering filing a formal complaint. If this issue cannot be resolved by early next week, we will discontinue adding new customers until it is fixed. Thanks for your understanding and continued support. -Dane

Tue Jan 6 20:29:04 PST 1998 — The repair issue mentioned in the last MOTD still has not been fixed, and we are experiencing busy conditions even though lines are available. As a short term solution, if you experience a busy signal, you should be able to dial past the bad area by calling 707-547-3210. This number will work for v.34, X2 and single-B ISDN.

I spoke with the FCC, the CPUC and Pacific Bell today, and it seems that there is an ISDN service guarantee that Pacific Bell filed with the CPUC last year which will allow us to get some credit for these lines which are not usable. I’ve gotten a bit of an agency run-around today, but we seem to be making progress. The folks at PacBell are working hard now on the repair, and they’ve escalated the issue with the technical support folks at Northern Telcom, manufacturers of the DMS100 switch which serves us. -Dane

Mon Jan 5 22:08:19 PST 1998 — Pacific Bell is still having capacity problems which are yielding busy signals for our customers. We’ve got 192 lines in one BRI based hunt group, and only the first 100 are usable. What this means is that we’ve got lines, and we’ve got modems, but users are getting a busy signal or reorder tone. We spent $45,000.00 on the equipment which serves these lines plus about $4600 in installation charges. Pacific Bell has been working on the issue since December 2nd. I spoke with the public utilities commission and the FCC today, and Sonic may be entitled to damages in this case in the amount of $230 per day, or about $7000 so far. We’ve informed Bell that we intend to persue this, and we’re hoping that they will escalate the repairs. We haven’t returned very many busy signals in the past few weeks to customers, but this Monday night has been particularly busy. See our lines graph for details: www.sonic.net/lines/