500 errors continue on web7. Stephen, have you gleaned any new info on this? It is becoming an constant problem that we surely do not what clients or their site visitors experiencing. We need to know what’s happening and if it is the system that can’t support common technology or what.
If the Linux system cannot run WordPress (an Easy App) that Jodo even promotes, how can we as resellers sell the service with any confidence?
This isn’t really my thing to glean info on, I don’t work on the linux systems except hardware level. There are many things being done here at JH, we just have had a few delays in getting it public yet, but it will be soon 
It will resolve the issues on web7, without a doubt, they are coming due to resource spikes, I know that much. We are going to ‘throw more resources’ at it as part of the solution.
Thought I’d chime back in and report that I received the 500 error on two diff sites today … one on web7 (twice) and one on web5. While these are currently in the dev phase, I’m very reluctant to release on Jodo servers given that this issue seems no closer to resolution.
I, and my design partner, run instances of WordPress sites, each on our own dev servers (one windows server and one Unix server) and we have yet to receive one issue with any of the several WP sites we currently run.
So, while I’m not certain how many of the Linux servers are experiencing this issue, but it’s more than just web7 and is a growing concern for me.
Stephen,
I just saw that you had responded as I sent my last message.
By “throw more resources” at it, what do you mean? Does this mean, add more memory/disk/etc. as more sites are added to a particular server, or perhaps add more servers?
Stephen - If the issue is resource spikes that would indicate a system problem…lack of resources or overloading. Either way, it isn’t an issue with “scripting” or “server abuse” which have been the explanations put forth on the support tickets I’ve posted since September (3 months) regarding this issue trying to get this resolved. Clearly it hasn’t and never has any answer been provided other than the one you just posted.
The problem for resellers & Jodohost: Our client sites are failing to perform as expected. Clients get upset when they discover this or are told by visitors. Resellers lose clients. Jodo loses clients because we will be forced to move these sites elsewhere due to inaction and pressure from clients who expect quality service. Moving enough sites for this reason…we might as well move them all.
If there is a solution that Jodohost KNOWS will solve the problem it might be a good idea to advise resellers of the plans, the progress, timeline and get it in place as soon as possible.
umm resource spikes can be due to abuse as well. In fact that is the common reason for a spike to occur, but we are working on multiple angles of this for long term fix across the board. I’ve expressed some of that in customer only for even higher usage sites as well.
Thanks Stephen. It seems that resource spikes of any kind should be able to be managed so that we are not experiencing this errors multiple times on a daily basis. If there are not enough resources to support the customers, that is a system design deficiency. I am happy that something is being done. But, this have been going on for at least 3 months on a regular basis. I can tell you my clients notice. We need a solution before another 3 months pass. Our clients simply will not tolerate it. It is bad for everyone’s business.
So, setting some expectations for resellers would be a nice change from “it’s working now” or “check it now” or “your scripts should be optimized” or “someone is abusing the server.” This is what the responses to my tickets have been. None of those are responses that help us manage our clients’ expectations. In fact, they make us sound evasive.
Here are two things we are doing on priority:
@squidz - agreed.
@tanmaya - thank you for the update!
Tanmaya - this new information is greatly appreciated. I am not versed in the fastcgi/mod_php PHP processing scenarios really, or how it affects the issue at hand, but if it improves the situation that is a positive.
Please keep resellers advised of progress and we will keep you advised of the performance results.
A new web server, and a MySQL server are available now. Since Hsphere doesn’t have MySQL move capability, we suggest to recreate account(s) if one wishes to make use of these new servers. Or a move can be requested, which will only move the web service.
I would also like to know how web7 and web8 has performed in past 24 hours?