Problems at start-up, runs slowly, then freezes.

Sounds like your boot drive is failing, if your RAID 0 is the boot, then you know the risks involved with running RAID 0 for anything.

If the RAID 0 is data, then simply get a new boot hard drive and install 10.5 on it or restore from a clone, leave the data RAID 0 alone, it will report to the new boot drive it's a set. Backup, backup, backup.

BTW, if it's a Intel Mac, you really shoudl be on 10.6.8, it's faster, more secure and a heck of a lot cooler than 10.5 (10.7 not so cool)

https://discussions.apple.com/thread/3529276

Disk Utility and 10.5, no.

Workng with it as is, no.

Not having a working clone to restore from...

:An SMC Reset, no, and 2008 not sure it had a botton, but wouldn't help.

The failure to boot Safe Mode says you, not Apple, need to just build or rebuild and maybe zero out, a new drive - leave as is what you have.

Even though the data array can stay, it too should have multiple backups and cloned.

Not finding a problem, no, just means those tests have limitations.

And there mat be things beyond a clean OS that factor in why it was sluggish, slow and not running normal.

may I assume that the disk is affirmatively, possitively OK

Disk Utility "Repair Disk" checks the integrity and consistency of the Directory, the "index" of what file is in what blocks on the drive. The entire rest of the drive could be non-functioning garbage and if the directory is OK, it would say, "no problems found".

As a drive ages, sometimes sectors develop problems, and incur re-trys when being read. If this gets bad enough, it will slow your Mac down. Sometimes this is reflected in the SMART status for the drive mechanism shown in Disk Utility (in the lower right when the Make & Model is selected), but often it is not.

A drive with a lot of data errors that cause re-trys can sometimes be "perked up" by re-writing all the blocks with zeroes. The drive mechanism will substitute spare blocks when this is done. But many suggest that a drive with read errors should be put into "light duty" service, such as backups, and eventually retired.

Tech Tool Pro has a "Surface Scan" function that exhaustively reads every block looking for read errors. Drive Genius has a similar feature. I expect there are othrs as well.

Disk Warrior finds more read errors and "Drive not responding" errors than Disk Utility, but does not do an exhaustive check.

TTPro though even though it does find weak or soft and bad sectors doesn't DO anything with them, whereas SMART Utility, a vendor's own, will. And Zero-all I call a blessing and prayer that it will. Micromat's own comment was that 7-way write was more reliable. My experience: even 7-way was not as helpful iwth a new WD drive but WD Lifeguard worked and drive has now had 4 yrs good use.

Hatter, you are quite right that those Utilities i mentioned do not FIX those errors. Their only use (in my opinion) is to diagnose WHETHER Bad Blocks were an issue on a particular drive. I have used them occasionally to determine whether slowdowns were caused by Bad Blocks, or whether I needed to be looking for other reasons.

that is why I like what SoftRAID has done by putting in a background media check - it always was more accurate to alert the user that there were I/O errors or 'weak' sectors - how many retries before a sector is mapped out? it isn't black and white.

The Google study that showed that using spares and diminished number of spares left was leading sign of drive failure could be impending.

any as long as it isn't SAS or something, overdue probaly too.

price hikes make it less attractive and won't improve soon.

SATA II/III 3.5"

Green, maybe, or not - they are less expensive good but not for primary data or system probably, due to delay spinups.