Status of Amtrak Status Maps - dixielandsoftware

[John Bobinyec]

Amtrak changed the size of the file that they send back from a status query. I check the size of the file and since they were suddenly smaller than a threshold that I use, they were being flagged as errors. I do this because when Amtrak refuses to answer a query, their response file is smaller than one where they actually give the information that I want. It's a method that I use to see if they are trying to shut me off. Anyway, they were providing the information okay, but because the files were a bit smaller, I thought there were problems. That effectively shut the system down. I was busy and didn't get to check and fix the problem until much later in the day.

jb

 

[Agent]

Some trains are still missing from the map. #6(12) for example.

 

[AmtrakLKL]

Glad it wasn't a major change. Back in my data analysis days, I had similar built-in checks to keep tabs on our IT department. They had a wonderful habit of changing things to suit them ("Oh, yeah. We ran an update last night, sorry."), but heaven forbid a user request changes. They'd mire you down in change requests forms, ROI spreadsheets and endless meetings to ensure your request didn't affect any other departments.

 

[John Bobinyec]

The Amtrak servers seem to be having problems this morning (10/12/2015), so that suspends Amtrak Status Maps, ASM Alerts and the Station Status Boards, until they come back up.

jb

 

[Palmetto]

Track-a-train doesn't want to open for several hours now either.

 

[montana mike]

The Amtrak servers seem to be having problems this morning (10/12/2015), so that suspends Amtrak Status Maps, ASM Alerts and the Station Status Boards, until they come back up.

jb

Still down as of this evening. Wow, in this day and age to have your entire web site down for this long is amazing. No info on any of the LD trains since midday.

 

[AmtrakBlue]

Sure, there's info out there, you just need to know who to ask. :giggle:

91 passed through WIL (Wilm, DE) around 1 pm. 92 passed through around 5 pm. 97 was running about 8 mins late to WIL (due in at 5:19 pm) the last time I checked (was going to wait for it to pass my SEPTA station but gave up as it lost more time).

 

[John Bobinyec]

If y'all check Amtrak.com right now (9.47 p.m. EDT, 10/12/2015) you'll see that Amtrak has lost their main website. They have put up a "lite" version of it so that you can still get some information and make simple reservations. Unfortunately, my system is incompatible with this one, so we'll just have to wait until they fix the main site. Hopefully it won't take too long.

jb

Amtrak Status Maps

ASM Alerts

Station Status Boards

 

[tim49424]

If y'all check Amtrak.com right now (9.47 p.m. EDT, 10/12/2015) you'll see that Amtrak has lost their main website. They have put up a "lite" version of it so that you can still get some information and make simple reservations. Unfortunately, my system is incompatible with this one, so we'll just have to wait until they fix the main site. Hopefully it won't take too long.

jb

Amtrak Status Maps

ASM Alerts

Station Status Boards

Amtrak's track-a-train still seems to be working. It just cannot be accessed from the main site.

 

[GG-1]

If y'all check Amtrak.com right now (9.47 p.m. EDT, 10/12/2015) you'll see that Amtrak has lost their main website. They have put up a "lite" version of it so that you can still get some information and make simple reservations. Unfortunately, my system is incompatible with this one, so we'll just have to wait until they fix the main site. Hopefully it won't take too long.

jb

Amtrak Status Maps

ASM Alerts

Station Status Boards

Very interesting, Yesterday news reported that due to a computer glitch the necessary boarding time was 4 hour at McCarren Airport in Las Vegas

Aloha

 

[montana mike]

Still down. The Amtrak IT guys need to be fired.

 

[Ryan]

Because we totally know that it's all their fault.

People that demand that someone gets fired without a clue as to what's going on should be shot. (See, I can make sweeping generalizations too!!!)

 

[montana mike]

OK. I have a wee bit of history and knowledge of this business, since that is what we do, and for hundreds of people, so I believe I can speak with some assurance here. In one way you have a point, since there could be some sinister third party that has struck and killed all of the backups they should have, et. al..... BUT with that being said web sites should not go down for days on end, especially not one of this magnitude. Have you ever seen Amazon or Google down for more than seconds or minutes? We have back ups and back ups for the back ups for ALL of our sites in at least three different locations in the US. If one of our server sites is totally destroyed, within a few seconds the first backup is up and running. The same is to be said if this is a really dumb coding error by someone trying to "update" the code over the weekend--which we never recommend or do btw (we beta test all updates off line first and more than once and then bring the new site up at one of our locations first. If that crashes, we immediately revert to the previous version and the screwed up version is pulled off line.

 

[Rail Freak]

OK. I have a wee bit of history and knowledge of this business, since that is what we do, and for hundreds of people, so I believe I can speak with some assurance here. In one way you have a point, since there could be some sinister third party that has struck and killed all of the backups they should have, et. al..... BUT with that being said web sites should not go down for days on end, especially not one of this magnitude. Have you ever seen Amazon or Google down for more than seconds or minutes? We have back ups and back ups for the back ups for ALL of our sites in at least three different locations in the US. If one of our server sites is totally destroyed, within a few seconds the first backup is up and running. The same is to be said if this is a really dumb coding error by someone trying to "update" the code over the weekend--which we never recommend or do btw (we beta test all updates off line first and more than once and then bring the new site up at one of our locations first. If that crashes, we immediately revert to the previous version and the screwed up version is pulled off line.

 

[Ryan]

Have you ever seen Amazon or Google down for more than seconds or minutes?

That's good that you're in the business - that means that you should have an even better appreciation of the fact that sometimes compost occurs.

Amazon had a somewhat severe issue with AWS here just recently as a matter of fact:

http://www.crn.com/news/cloud/300077740/overnight-aws-outage-reminds-world-how-important-aws-stability-really-is.htm

As did Google Docs last week:

http://www.techtimes.com/articles/93822/20151011/google-docs-outage-was-the-perfect-excuse-for-workers-to-slack-around-and-create-these-hilarious-memes.htm

They even provide a dashboard to let you know what's up and what's down at any given time:

http://www.google.com/appsstatus#hl=en&v=status

So yeah, even when best practices are follows, sometimes the fecal matter comes into contact with the rotating blades, which makes sitting back from afar without any knowledge of the specific situation pretty dumb.

 

[montana mike]

dumb answer!!!

I try to explain how things should NOT happen if they are properly set up and all you spew is stuff about "fecal matter", as if web sites go down for NO reasons at all. Unless you are in the business, as we are and have been for 11 years, this should not have happened-period. Yes, sites go down for very short time frames, but we are now closing in on 48 hours for a major e-commerce site that likely has hundreds of thousand of hits each day.

S**t does not happen if it is done right.

 

[Ryan]

You asked about Google and Amazon being down and I provided recent examples. I'd be a little upset too if my point were deconstructed so completely.

If only we were all as perfect as you are, the trains would run on time and the website would be perpetually available.

**** always happens. Always.

 

[jebr]

Let's ensure that the discussion does not degenerate into personal attacks. Thanks.

 

[Phil S]

10:24 PST. Amtrak.com dysfunctional. Can't make reservations. I don't see how to get to Track A Train. OTOH, the alternative track a train works fine. Amtrak Status Map (jb Dixieland) seems to be working fine. Sorry, I have to agree with Mike - this should not be happening, even given the extreme complexity of what Amtrak.com does (reservations, award stuff, system status map, etc). Just one opinion!

 

[NW cannonball]

Amtrak.com seems back to normal now. I had to clear my firefox cache (and delete some cookies just for good measure) and I have no current reservations, so I can't test many functions,

But it seems to be back up and running.

As for the dixielandsoftware status maps, the official datafeed only came online an hour or 4 ago, some of the Western LD trains that departed during the downtime are not yet showing there , but mostly OK. Thanks, John

Glad that disruption is over with.

Take care, take trains.

 

[Ryan]

Sorry, I have to agree with Mike - this should not be happening

I don't think that anyone is arguing that it should be happening, I am most definitely not.

My sole objection is the knee jerk "Fire the IT people!" cry when we have no idea why the site was down and who (or what) was actually at fault.

 

[John Bobinyec]

When budgets/schedules are tight, the first thing that goes is documentation. The second thing is backups.

jb

 

[xyzzy]

When budgets/schedules are tight, the first thing that goes is documentation. The second thing is backups.

Third, fourth, fifth, and sixth are code reviews, formal testing, formal change control/config management, and escrow deposits of source code... not necessarily in that order. Have seen it happen too many times.

 

[neroden]

At some point, informal testing goes too.

 

[John Bobinyec]

Another one of life's little disappointments. The emailer program that sends out the Alerts is not working. The code hasn't been touched in years. And it looks normal. I suspect there's trouble with my ISP's SMTP server, but I have yet to pin it down.

jb