Jump to content

Website Problems Yet Again?

The Cache Couple

Recommended Posts

I don't want to sound like a crying baby, but can anyone explain to me when the site will be fixed once and for all? I can understand little problems from time to time, but if I am not getting errors trying to access the site, then I can't get my pocket queries. It seems to happen on a regular basis.


It doesn't seem to be getting better. For several months we have been experiencing problems and they go away only to return a day or so later. This is not a bashing topic, yet one of concern. There are numerous folks in my caching area that express the same concerns or frustrations.


Can anyone explain when the above concerns will be fixed? Thanks for your reply and again, this is not a bashing topic. Thanks!

Link to comment
I don't want to sound like a crying baby, but can anyone explain to me when the site will be fixed once and for all? 

It can only be "fixed" when the sport stops growing and evolving. Personally, I hope it never reaches that stage.

I don't think that's how Yahoo! and Google handled their growth, and I don't think this is what you meant, but it comes off like: "As long as our business is growing, the customers will have problems because we can't keep up. Only when the growth rate in demand falls below our capacity growth rate will our customers be happy."


Seriously, we know work is being done to improve performance to be able to handle the growth in users and the hobby/sport.


I, for one, appreciate the difficulty of the task in front of GC.COM, because I work with large/complex databases all the time, and performance tuning is a very important issue when databases become extremely large/complex, especially when viewing that data through different parameter filters such as user preferences (i.e. found/unfound). When the detail becomes extremely large, the processing of these types of joins can be extremely expensive. And I see all these platform flame wars, which amuse me, since none of these people have any idea that this technology problem is not unique to any one platform - you would have ALL THE SAME PROBLEMS on every platform when a database gets to this state of size, complexity, etc. It seems like you run into every problem all at once - table size, number of tables, number of joins, clustered indexes which used to be right for data at one size, but not so good when the data evolves over time.


I think the question, reasonably asked and answered, which most of the frustrated customers probably have, is simply:


We know you're working hard on the problems, but when do you think the site will be able to handle a weekend again?

Link to comment

We know you're working hard on the problems, but when do you think the site will be able to handle a weekend again?


I agree. I appreciate the work that has been done to improve the site over the last few years. I also appreciate the magnitude of the challenge. I would however like to know what steps are being taken and when we can expect conditions to improve.

Things often look better when you have a timeframe for when you can expect to see relief.

Link to comment

This is purely speculation but maybe they don't say "We hope to have problem 'x' fixed within the week".


I know from experience with other things if you give people a certain date and things aren't in place on that exact date you'll get everyone asking why isn't x fixed yet?

Link to comment
Oh well, I guess I'll check the pages of a couple local caches for any changes, then go out with old data.

Tried that once altely, I love looking for archived caches! LOL :blink:

That's why I said "check the pages for changes" :blink:


As long as nothing on the cache page changed, my PQ was "current enough".

Link to comment

Why is everyone jumping to the conclusion that this is due to too many users using the database and/or the PQ service? That certainly is not the only possibility. In fact, we know that the PQs run on a separate machine on a copy of the database, and that it is normally finished processing the scheduled PQs fairly early in its 24 hour cycle. So there are plenty of spare clock-cycles in that operation.


Here are some other potential culprits: Maybe the firm that provides the internet service now has too much traffic on its resources; Maybe there is some hardware defect in the server, router or firewall (e.g. cables, connectors, NIC's etc go bad); Maybe someone hacked into the server and is using it to serve out MP3s or similar (it happened to one of my machines); I suspect many of us here could list many more possibilities.


Jeremy - What say you? What tests have you run? What's the through-put capacity and how close are we to it at various times. What's the max number of users the system can handle at any one time, and why are you letting them all on, only to all get lousy service. Let's see the dirty laundry.


Just now even the Forum is timing out occaisionally - that's silly - you don't need a lot of hamsters for that service even if hundreds of folks are on (153 are on-line right now). Seems like someone needs to check on whether GC's ISP is short-changing geocaching.

Link to comment
This topic is now closed to further replies.
  • Create New...