Yellow banner alert

7.3k · October 30, 2012

:omnomnom:

Geocaching.com will be going offline temporarily for a site update on Tuesday, October 30, 2012 at approximately 6pm PDT (convert to your local time here) for about 2 hours. During this time there will be no access to services including the Geocaching.com website, Geocaching phone applications, and Waymarking.com.

:omnomnom:

B.

Edited October 30, 2012 by Pup Patrol

8.3k · October 30, 2012

Oh, playing with the hardware again.

7.8k · October 30, 2012

October 30, 2012

You know, I really don't understand why Geocaching.com has to EVER go down. I understand that servers have to be maintained, or software has to be modified.

But why on EARTH don't you have a backup server that you put on line, do the work on the main server, then switch back? It seems so simple.

Google NEVER goes down for "maintenance". Why should gc.com? You'd think a company worth over 2 million dollars could afford a backup server.

Edited October 30, 2012 by jimlips

3.6k · October 30, 2012

Wonder if it got something to do with this? <_<

8.3k · October 30, 2012

Wonder if it got something to do with this?

Don't think so. The software updates are usually done in the morning/early afternoon and the hardware is done in the evening.

8.3k · October 30, 2012

You know, I really don't understand why Geocaching.com has to EVER go down. I understand that servers have to be maintained, or software has to be modified.

But why on EARTH don't you have a backup server that you put on line, do the work on the main server, then switch back? It seems so simple.

Google NEVER goes down for "maintenance". Why should gc.com? You'd think a company worth over 2 million dollars could afford a backup server.

First off there is more than one server involved in geocaching.com. A simple swap of a server won't work. If they are reconfiguring networking or fiber cables for the site that will probably require the site going down. If they are doing a wholesale upgrade of hardware that will require the site going down. Hey it is on Tuesday, the slowest day of the week for them.

Google is hundreds of servers on multiple sites with a much bigger bag of money than Groundspeak. How did you come up with the net worth? It is a private company and they tend not to share their financials.

19k · October 30, 2012

Geocaching.com will be going offline temporarily for a site update on Tuesday, October 30, 2012 at approximately 6pm PDT (convert to your local time here) for about 2 hours. During this time there will be no access to services including the Geocaching.com website, Geocaching phone applications, and Waymarking.com.

B.

I remember the outage we had when they gave us Donut Signal. That may well have been the best downtime ever.

7.7k · October 30, 2012

Google NEVER goes down for "maintenance". Why should gc.com? You'd think a company worth over 2 million dollars could afford a backup server.

Assuming Groundspeak is worth $2 million, you think comparing a $2 million company to a $200 billion company makes perfect sense?

7.7k · October 30, 2012

I remember the outage we had when they gave us Donut Signal. That may well have been the best downtime ever.

I don't know, I think the coffee that goes with the donuts were a little on the weak side.

14k · October 30, 2012

I remember the outage we had when they gave us Donut Signal. That may well have been the best downtime ever.

I don't know, I think the coffee that goes with the donuts were a little on the weak side.

That's only because the hamsters are hogging all the strong stuff.

1.1k · October 30, 2012

2 hrs with no online live caching..

and no loging..

I survive fine..

October 30, 2012

I would guess the difference between Google and GC is that Google probably has many different paths to get you where you want to go and GC probably has very limited paths and therefore can't re-direct you. Google servers come down like everyone elses but because they can re-direct you it isn't obvious to you.

15.3k · October 30, 2012

You know, I really don't understand why Geocaching.com has to EVER go down. I understand that servers have to be maintained, or software has to be modified.

But why on EARTH don't you have a backup server that you put on line, do the work on the main server, then switch back? It seems so simple.

For a single server, it can be that simple.

When you build a web site that uses multiple servers that interact in various ways, it's a bit more complicated. It can still be done, but it's more complicated. For example, you can pull half the systems out of the load balancer, wait for the connections to expire, and then update those systems. Then you swap the first group of systems (updated) and the second group of systems (not yet updated) in the load balancer, at which point your update is live. Then you update the second group of systems and add them back to the load balancer. This assumes that you have redundant systems for everything, that half your systems can handle the load during the upgrade, that you've choreographed the update so everything goes smoothly, etc. But it can be done.

Google NEVER goes down for "maintenance". Why should gc.com? You'd think a company worth over 2 million dollars could afford a backup server.

I used to work at Google. They operate at an entirely different level. In a sense, some part of Google is ALWAYS "down for maintenance". But that's how their infrastructure is designed to work. The EDS "Airplane" commercial isn't that far off:

7.5k · October 30, 2012

You know, I really don't understand why Geocaching.com has to EVER go down. I understand that servers have to be maintained, or software has to be modified.

But why on EARTH don't you have a backup server that you put on line, do the work on the main server, then switch back? It seems so simple.

Google NEVER goes down for "maintenance". Why should gc.com? You'd think a company worth over 2 million dollars could afford a backup server.

Other responses have objected to the google comparison as being the wrong order of magnitude, but that's not really relevant. The real answer is that while it seems simple, and it is relatively straightforward, it does still require resources in development, planning, and hardware. I'm reasonably confident that the people maintaining geocaching.com are well aware of the possibility, but for now they've decided that an occasional 2 hours outage is not a significant enough problem to worry about and decided to allocate their resources elsewhere.

Although another possibility is that they're already perfectly capable of doing this, they just don't because they want to make sure that for at least 2 hours a month, people actually go out and look for caches.

19k · October 30, 2012

I remember the outage we had when they gave us Donut Signal. That may well have been the best downtime ever.

I don't know, I think the coffee that goes with the donuts were a little on the weak side.

Maybe that's what this upgrade is all about. Fresh coffee. Seattle is all about fresh coffee.

20.7k · October 30, 2012

I'm not an IT guy, but it seems to me that the google comperison is off because the main google site doesn't actually 'do' anything. They could have five thousand identical servers doing exactly the same thing and take them up and down at will and it doesn't matter as long as when you google your name stuff comes up. That google search doesn't absolutely have to be saved for posterity like a cache log does.

It should be noted that the part of google that does need to save things (I'm looking at you, Gmail) does go down from time to time.

2.3k · October 30, 2012

Somehow I really doubt that Groundspeak's server farm looks anything like Google's.

--Larry

1.1k · October 31, 2012

19 minutes and counting. :laughing:

14k · October 31, 2012

I'm not an IT guy, but it seems to me that the google comperison is off because the main google site doesn't actually 'do' anything. They could have five thousand identical servers doing exactly the same thing and take them up and down at will and it doesn't matter as long as when you google your name stuff comes up. That google search doesn't absolutely have to be saved for posterity like a cache log does.

It should be noted that the part of google that does need to save things (I'm looking at you, Gmail) does go down from time to time.

You're missing an important point about how's actually searched is an index (a very large index) that is maintained on their server and constantly updated and probably replicated in many different physical locations. The user interface portion of google search may be minimal but the search index has be available

October 31, 2012

it's been more then 2 hours lol

2.8k · October 31, 2012

Hmmm, I couldn't log in from work to the forums or on my cell phone, I only got on at home where I was already logged in.

7.8k · October 31, 2012

it's been more then 2 hours lol

It's now 3 1/4 hours, and the ever-helpful ( :laughing: ) Geocaching System Status page says everything is just hunky-dory.

<_< :rolleyes: :unsure: :blink: :huh:

8.3k · October 31, 2012

it's been more then 2 hours lol

It's now 3 1/4 hours, and the ever-helpful ( ) Geocaching System Status page says everything is just hunky-dory.

It is, the hamsters are quite happy and running in their exercise wheels.

October 31, 2012

Going on 4 1/2 hours now :cry:

5.6k · October 31, 2012

It's up now.

20.7k · October 31, 2012

I'm not an IT guy, but it seems to me that the google comperison is off because the main google site doesn't actually 'do' anything. They could have five thousand identical servers doing exactly the same thing and take them up and down at will and it doesn't matter as long as when you google your name stuff comes up. That google search doesn't absolutely have to be saved for posterity like a cache log does.

It should be noted that the part of google that does need to save things (I'm looking at you, Gmail) does go down from time to time.

You're missing an important point about how's actually searched is an index (a very large index) that is maintained on their server and constantly updated and probably replicated in many different physical locations. The user interface portion of google search may be minimal but the search index has be available

That doesn't really change my point. If Google's backup server's index is a bit stale, it doesn't matter.

14k · October 31, 2012

I'm not an IT guy, but it seems to me that the google comperison is off because the main google site doesn't actually 'do' anything. They could have five thousand identical servers doing exactly the same thing and take them up and down at will and it doesn't matter as long as when you google your name stuff comes up. That google search doesn't absolutely have to be saved for posterity like a cache log does.

It should be noted that the part of google that does need to save things (I'm looking at you, Gmail) does go down from time to time.

You're missing an important point about how's actually searched is an index (a very large index) that is maintained on their server and constantly updated and probably replicated in many different physical locations. The user interface portion of google search may be minimal but the search index has be available

That doesn't really change my point. If Google's backup server's index is a bit stale, it doesn't matter.

Unless you're looking for some recent time sensitive information. Frankly, getting current crisis related information seems to me to be a lot more important than checking to see a Found It log was posted on a cache in the last two hours.

20.7k · October 31, 2012

I'm not an IT guy, but it seems to me that the google comperison is off because the main google site doesn't actually 'do' anything. They could have five thousand identical servers doing exactly the same thing and take them up and down at will and it doesn't matter as long as when you google your name stuff comes up. That google search doesn't absolutely have to be saved for posterity like a cache log does.

It should be noted that the part of google that does need to save things (I'm looking at you, Gmail) does go down from time to time.

You're missing an important point about how's actually searched is an index (a very large index) that is maintained on their server and constantly updated and probably replicated in many different physical locations. The user interface portion of google search may be minimal but the search index has be available

That doesn't really change my point. If Google's backup server's index is a bit stale, it doesn't matter.

Unless you're looking for some recent time sensitive information. Frankly, getting current crisis related information seems to me to be a lot more important than checking to see a Found It log was posted on a cache in the last two hours.

... and then we argue for twenty more posts about junk that doesn't matter even in the context of the OP.

Whatever.

3k · October 31, 2012

I guess the implementation of new features is being delayed by the need to upgrade/fix hardware?

8.3k · October 31, 2012

I guess the implementation of new features is being delayed by the need to upgrade/fix hardware?

Guess you didn't read what Moun10bike said ...

Last night's downtime was for a hardware update. We no longer have to take the site down fro software updates, and since much of the work done each sprint is infrastructure work and bug fixes, we have elected to not post release notes unless an update involves a major new feature.

I assume that the hardware updates were such that service could not be maintained during the update.

3.6k · October 31, 2012

I guess the implementation of new features is being delayed by the need to upgrade/fix hardware?

Guess you didn't read what Moun10bike said ...

Last night's downtime was for a hardware update. We no longer have to take the site down fro software updates, and since much of the work done each sprint is infrastructure work and bug fixes, we have elected to not post release notes unless an update involves a major new feature.

I assume that the hardware updates were such that service could not be maintained during the update.

So that mean they wont tell us what they updated or change or fixed?

8.3k · October 31, 2012

I guess the implementation of new features is being delayed by the need to upgrade/fix hardware?

Guess you didn't read what Moun10bike said ...

Last night's downtime was for a hardware update. We no longer have to take the site down fro software updates, and since much of the work done each sprint is infrastructure work and bug fixes, we have elected to not post release notes unless an update involves a major new feature.

I assume that the hardware updates were such that service could not be maintained during the update.

So that mean they wont tell us what they updated or change or fixed?

Is that what he said? I was wondering about that.

5.7k · October 31, 2012

I took it as three separate statements.

1. We didn't update the software last night, just hardware.

2. We don't have to take the site down anymore to do software updates.

3. Lately all the software updates we've done have been boring, so rather than post updates every time we do that, we're holding off on announcing updates unless we roll out a major new feature.

Begging the question, will we see updates if they fix a MAGOR BUG?

7.8k · October 31, 2012

Begging the question, will we see updates if they fix a MAGOR BUG?

I'd like to see descriptions of any bugs that get fixed, major or minor. Some bugs have been very long-standing, so it would be nice to know if/when these get fixed rather than having to constantly test it ourselves to see if the bug has been fixed.

October 31, 2012

I took it as three separate statements.

1. We didn't update the software last night, just hardware.

2. We don't have to take the site down anymore to do software updates.

3. Lately all the software updates we've done have been boring, so rather than post updates every time we do that, we're holding off on announcing updates unless we roll out a major new feature.

Begging the question, will we see updates if they fix a MAGOR BUG?

At least the server sw-build was updated from : web.hotfix_20121016.3 (?) to web.hotfix_20121030.1.

Whatever the content of the fix???

19k · October 31, 2012

I guess the implementation of new features is being delayed by the need to upgrade/fix hardware?

Guess you didn't read what Moun10bike said ...

Last night's downtime was for a hardware update. We no longer have to take the site down fro software updates, and since much of the work done each sprint is infrastructure work and bug fixes, we have elected to not post release notes unless an update involves a major new feature.

I assume that the hardware updates were such that service could not be maintained during the update.

So that mean they wont tell us what they updated or change or fixed?

Is that what he said? I was wondering about that.

unless an update involves a major new feature.

3.6k · October 31, 2012

I guess the implementation of new features is being delayed by the need to upgrade/fix hardware?

Guess you didn't read what Moun10bike said ...

Last night's downtime was for a hardware update. We no longer have to take the site down fro software updates, and since much of the work done each sprint is infrastructure work and bug fixes, we have elected to not post release notes unless an update involves a major new feature.

I assume that the hardware updates were such that service could not be maintained during the update.

So that mean they wont tell us what they updated or change or fixed?

Is that what he said? I was wondering about that.

unless an update involves a major new feature.

Thats how I read it as well. I feel its a wrong move on GS part. Its important to let your customers know of any fixed you made. I know every fixed upgrade of my computer games, software or not.

Not so long ago, GS made a change on their API server and everybody was wondering whats going on. What happen was that GS changed a setting. If GS was open about every changes or fixed they made, this confusion would never happen. I am not going to debate about the changes they made here, I am here talking about lack of communication with their customers.

5k · November 1, 2012

You know, I really don't understand why Geocaching.com has to EVER go down. I understand that servers have to be maintained, or software has to be modified.

But why on EARTH don't you have a backup server that you put on line, do the work on the main server, then switch back? It seems so simple.

Google NEVER goes down for "maintenance". Why should gc.com? You'd think a company worth over 2 million dollars could afford a backup server.

Really? Google has over 900,000 servers.

I tried to pay my cable bill the other night and Time Warner's web site was down for maintenance. You would think that a company worth 35 billion dollars could afford backup servers.

14k · November 1, 2012

Begging the question, will we see updates if they fix a MAGOR BUG?

I'd like to see descriptions of any bugs that get fixed, major or minor. Some bugs have been very long-standing, so it would be nice to know if/when these get fixed rather than having to constantly test it ourselves to see if the bug has been fixed.

+1

We've got the Feature suggestion forum where lots of people have suggesting things that should done and some of them would require rather minor changes. If GS isn't responded to indicate whether not a feature might get implemented and we're not told if/when a bug has been fixed it's leading to the perception that GS isn't anything to improve it's service to their customers.

7.8k · November 1, 2012

We've got the Feature suggestion forum where lots of people have suggesting things that should done and some of them would require rather minor changes. If GS isn't responded to indicate whether not a feature might get implemented and we're not told if/when a bug has been fixed it's leading to the perception that GS isn't anything to improve it's service to their customers.

Just today, over in the Bug Report forum, a user asked what the status was for a bug. Moun10Bike responded that it had already been fixed. Unfortunately, until he posted that, there was no way any of us would know that bug had been fixed other than repeatedly trying it ourselves.

14k · November 2, 2012

We've got the Feature suggestion forum where lots of people have suggesting things that should done and some of them would require rather minor changes. If GS isn't responded to indicate whether not a feature might get implemented and we're not told if/when a bug has been fixed it's leading to the perception that GS isn't anything to improve it's service to their customers.

Just today, over in the Bug Report forum, a user asked what the status was for a bug. Moun10Bike responded that it had already been fixed. Unfortunately, until he posted that, there was no way any of us would know that bug had been fixed other than repeatedly trying it ourselves.

Where I work we use a well known issue tracking system. When someone reports a bug, it's assigned to someone whereupon it goes through several stages. Once the developer has fixed it, the change is deployed to a test system and the developer marks it as resolved. Only after the issue has been confirmed to be fixed by the user that reported it is the issue closed. This is nothing new for me. I was essentially following the same procedures when I was responsible for bug fixing and enhancements on a proprietary operating system in the early 1980s. It's amazing that a company that develops software with over a million customers in 2012 won't keep it's customers notified about the state of their software.

7.8k · November 2, 2012

Where I work we use a well known issue tracking system. When someone reports a bug, it's assigned to someone whereupon it goes through several stages. Once the developer has fixed it, the change is deployed to a test system and the developer marks it as resolved. Only after the issue has been confirmed to be fixed by the user that reported it is the issue closed.

That's basically how it works in our IT section. When someone comes to us with a problem, it goes into the tracking system. Each ticket doesn't get closed until the user has confirmed that the problem has been fixed. I would never even consider closing a ticket if I wasn't sure that the user was satisfied that the problem had indeed been resolved, or at least worked-around to their satisfaction.

5k · November 2, 2012

We've got the Feature suggestion forum where lots of people have suggesting things that should done and some of them would require rather minor changes. If GS isn't responded to indicate whether not a feature might get implemented and we're not told if/when a bug has been fixed it's leading to the perception that GS isn't anything to improve it's service to their customers.

Just today, over in the Bug Report forum, a user asked what the status was for a bug. Moun10Bike responded that it had already been fixed. Unfortunately, until he posted that, there was no way any of us would know that bug had been fixed other than repeatedly trying it ourselves.

It seems that in about a years time Groundspeak has gone from total transparency with open feedback forums to an almost cloak and dagger type attitude. It's too bad that they couldn't have found some middle ground. Even a locked "release notes" type post would be helpful.

7.8k · November 2, 2012

Even a locked "release notes" type post would be helpful.

That would be more than enough for me. I always used to read through every bullet point in the release note posts, seeing which bugs had been fixed and what things had changed. With the way it is now, figuring out what has changed is like searching for an evil micro.

3.6k · November 2, 2012

Even a locked "release notes" type post would be helpful.

That would be more than enough for me. I always used to read through every bullet point in the release note posts, seeing which bugs had been fixed and what things had changed. With the way it is now, figuring out what has changed is like searching for an evil micro.

Same here. I always go down the list even its locked to see what they had worked on.

5.2k · November 2, 2012

Where I work we use a well known issue tracking system. When someone reports a bug, it's assigned to someone whereupon it goes through several stages. Once the developer has fixed it, the change is deployed to a test system and the developer marks it as resolved. Only after the issue has been confirmed to be fixed by the user that reported it is the issue closed.

That's basically how it works in our IT section. When someone comes to us with a problem, it goes into the tracking system. Each ticket doesn't get closed until the user has confirmed that the problem has been fixed. I would never even consider closing a ticket if I wasn't sure that the user was satisfied that the problem had indeed been resolved, or at least worked-around to their satisfaction.

I hear that Groundspeak uses little slips of paper with the problems written on them, all thrown into a fishbowl. A programmer randomly fishes out a slip of paper, fixes the problem, then throws the paper away. If the programmer determines the problem can't be fixed at that time, he puts it in another fishbowl.

The only real problem with this method is occasionally someone forgets to feed the fish and they start eating the slips of paper. :omnomnom: :drama:

3.6k · November 2, 2012

Where I work we use a well known issue tracking system. When someone reports a bug, it's assigned to someone whereupon it goes through several stages. Once the developer has fixed it, the change is deployed to a test system and the developer marks it as resolved. Only after the issue has been confirmed to be fixed by the user that reported it is the issue closed.

That's basically how it works in our IT section. When someone comes to us with a problem, it goes into the tracking system. Each ticket doesn't get closed until the user has confirmed that the problem has been fixed. I would never even consider closing a ticket if I wasn't sure that the user was satisfied that the problem had indeed been resolved, or at least worked-around to their satisfaction.

I hear that Groundspeak uses little slips of paper with the problems written on them, all thrown into a fishbowl. A programmer randomly fishes out a slip of paper, fixes the problem, then throws the paper away. If the programmer determines the problem can't be fixed at that time, he puts it in another fishbowl.

The only real problem with this method is occasionally someone forgets to feed the fish and they start eating the slips of paper.

Coming from you, I am shocked. :blink:

7.8k · November 2, 2012

The only real problem with this method is occasionally someone forgets to feed the fish and they start eating the slips of paper.

Well, they are bugs, aren't they?

14k · November 2, 2012

The only real problem with this method is occasionally someone forgets to feed the fish and they start eating the slips of paper.

Well, they are bugs, aren't they?

<rimshot>

Know else eats bugs? Frogs.

Yellow banner alert

Recommended Posts

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment