0

A client with a small server was running out of disk space. It is Dell PowerEdge 830 with 2 GB RAM running under Windows 2003 Standard.
The machine came from Dell with three 72GB Seagate (Cheetah 10K.7, 10K RPM, ST373207LW Ultra U320) SCSI drives which Windows saw as a 15GB C: drive and 125 GB D: drive. (The drives are attached to a PERC 4/SC RAID controller.)
Instead of replacing the existing SCSI (RAID) array I decided to just add a second array with three more Seagate 300GB SCSI drives. (I can't believe I did not log their model number!) That went fine. Windows now showed a new 600 GB drive. It partitioned and formatted uneventfully.
I then copied everything in the D: drive to the new one, deleted the old D: drive, and made the new 300 GB volume the "new D:".
Shortly after that, the client reported that when they called up some (graphic) documents (like faxes) in their application they experienced random delays of 5, 10, sometimes 15 seconds.
They found that the delays go away for about a day or so if they do a defragmentation. I personally think defragmentation is one of the most overrated operations I know and it cannot be a direct factor in an operation with only 10 or 12 terminals which does NOT expand many files. Mainly, they create NEW ones.
So, I think the defragmentation cure is actually a "misidentification". I think Defrag is doing something else (it reports 0 fragmentation when launched) that alliviates the problem but I cannot figure out what.
Basically, when they are looking at a screen, they select some fax or image they want to see and it sometimes "stalls" the application for 5, 10, or even 15 seconds in that they must wait that long to see the image. (The application happens to be eClinical Works, a program to run a paperless medical clinic.)

It smells like perhaps swapping activity in the (dedicated) Server but . . . What's going on here??

1
Contributor
1
Reply
2
Views
7 Years
Discussion Span
Last Post by CarlosCoquet
0

I checked the disks I added: they are 3 ST3300007LW. The original 3 disks are at addresses 0, 1, and 2 of the controller's Channel 1. The ones I added are at addresses 8, 9, and 10 of the same. (I checked each drive's stat's and they all said 0 errors.)
The machine is running TomCat 5 but I am pretty sure it is only for "internal" purposes. For example, some eClinical Works operations, like installing the program in a new workstation, address port 8080 on the Server's IP address.
Another piece of the puzzle is MySQL. I think that is what eClinic Wroks uses for its data.

This topic has been dead for over six months. Start a new discussion instead.
Have something to contribute to this discussion? Please be thoughtful, detailed and courteous, and be sure to adhere to our posting rules.