Quote:
Originally Posted by Crion
Day 2 of yet another epic peq phailure.
/scratches
I need mah crack!
&& in case you havn't noticed, each time the server goes down we are losing more players to other servers, like the EZ server(!!!?) and worse yet, p1999 .. Sadly I can even name a few people that are gone for good..
I understand to you this may be an important time to ask for donates, but you said yourself the current set up is already more than enough for now.
Why so much downtime and so frequent? There are other servers out there with far less equipment and bandwidth and yet they have longer up times, some only reset once a week, if that, and with roughly the same player load.
I'm not knocking the need for new server upgrades by any means, I'm just wondering how or why PEQ seems to have poor reliability with its current set up, or rather if the problem is the ISP/Host, why continue to subscribe to their services if they are so incompetent.
|
The hosting is being provided for free by the hosting company that FNW used to work for. He no longer works there and his new job requires him to travel frequently so cavedude is not always able to contact him immediately when there is a need.
In this case, cavedude did contact FNW and FNW put in a reboot request with the hosting company. But since they are providing the service for free, clearly this is not going to be a priority for them.
The current machine is adequate for our needs. However, it does not have much available headroom so if there is a bug in the server code that causes heavier load than normal, we will experience problems. Even more than that, with the current set up, we do not have remote management capabilities so we are completely at the mercy of FNW and the hosting company when we do have problems that cause the server to be unreachable.
To resolve all of the above problems, we would like to purchase two new systems that will exceed our needs by a wide enough margin we should not see performance problems from a player perspective even if there are issues with the server code. These systems will include full remote management capabilities allowing us to deal with nearly all problems remotely. And the fact that there will be two will allow us to have a failover in the event of one system failing completely. This should all add up to no downtime or miniscule downtime.
Some people ask about upgrading the existing system to accommodate these features. The problem with that is how long do you think it would take them to upgrade our system when it takes them days to simply reboot it? And that is assuming they would even be willing to do the upgrade for us, which they may very well not be willing to do.
Next we get asked about moving the hosting to a different location. Again it comes down to how long it would take the hosting guys to pack up and ship the system, which isn't very likely to happen in a timely manner either. And again, that assumes they are even willing to do so. They might require it to be picked up, which would require FNW to do so, which he may not have the time for.
So the solution is new servers with remote management at a hosting location that multiple people can contact. The remote management will be available to KLS, cavedude and myself. The hosting location could be called directly by the same 3 people. If physical access is required, it is a 25 minute drive for me. But it is not very likely physical access will be needed with the remote management capabilities.
But all of this costs money. We are doing what we can with the resources we have. We will improve things as we have the resources to do so. Until then, be patient and donate if you are so inclined.