Hi All,

Got a bunch of short-cuts for quick access to project status filters with valid and pending and error, return date sorted, some published in the Results Status page FAQ and all work top notch performance for me.

I wonder, my RS page total, which I extract using Office predefined links, showed 19 pages yesterday with older stuff. Today I have 22 pages. Maybe the performance goes digressively down with more pages, so how many have you?

[Edit: To quickly find out with many, hit the Last link. Got a ludicrous number in the shortcut, but the number that appears left of "Last" is your number]

I don't understand, nor did I get an answer to the pages question.

I have a bunch of shortcuts, only few are in the FAQ. Most work recently was FAAH, so when I place a filter on Valid/FAAH or Pending Validation / FAAH and even PV / FAAH / Device XXX, it remains top notch here. Now I'm on a good internet connection too using FFX 3.0.

Anyway, to emulate yours, i filtered on All / FAAH and after 2 seconds had 13 pages listed. Filtered FAAH on "In Progress" and it took 3 seconds. Hit the Last page link and it took 5 seconds, so cant regrettably repeat your experience .... need more volume information from your side.

Remember, this is real time direct hit on the Result Status tables that are continuously updated by the Scheduler.... yesterday there were 265 results or 3 results per second coming in and same going out and In Progress filtering is I think the most volatile value. I'm in the backwaters 4km from the nearest switching station, yet have a HQ service.

I get even team member listings for the very large, sometimes it needing a F5, so wonder if the ISP speed has anything to do with this. Getting a time out takes a very long time, here (may have tweaked it somewhere).

the servers are big time upgraded as is the db version and all moved to a better hosting location that remains accessible even during blizzards. This to happen in September/October, so don't think much can be done this moment.

Off topic, I switched, unbeknownst to a misconfiguration, a Wifi dongle from USB 1.0 to a USB 2.0 port on 1 device. MS in the August updates sneaked in a few reliability patches and a screen popped up yesterday to suggest I do this (was very impressed it did this). The whole local network (only WIFI), not a single wire, went in blazing mode. File queries on the wlan are next to instantaneous now.

Sekerob,

I did read the your last 2 posts, and I apologize for not responding. I was kind hoping that the problem would go away on it's own and I could say that it was [Resolved].

Right now I can't get any "Results Status", but it was working a few minutes ago, at that time I had a total of 40 pages of results.

My Internet connection, according my ISP's speed test, is running at 20mbps down and 600kbps up.

I'll try to explain better what I'm seeing:

When I go into the initial "Results Status" page it usually takes about 1 second or less to load. click any page number or last is pretty much instantaneous.

If I filter All, All, In Progress, it's still OK. The problem only occurs when I use the "In Progress" and the "Project Name" filter at the same time.

Having said all that, the problem seems totally random. I can't reproduce it at will.

Can we wait until after the server move/upgrade or would you like to just put it down to database contention?

Thanks.

Pete

Sekerob,

I have to apologize again, I run run ALL WCG projects.



This could be any project.

Currently I have 2 pages of FAAH (17 jobs), 1 in progress, I don't think it's a volume thing.

The 40 pages I referred to before is ALL projects and ALL result states.

Pete

Ok - it looks like in some cases, MySQL is using a less than efficient index. We are adding some additional clauses to the queries so that they will hit better indexes which should resolve the problem. These will be part of the next website build.