OT: Linked Out?

I had noticed that with the "4V" drivers. Since the Li-Ion chargers were 3.6V, I figured the marketeering departments were in power.

Reply to
krw
Loading thread data ...

formatting link

recommend

formatting link

Kinda like my framing nailer. ;-)

formatting link

16d Framing nailers aren't toys either.
Reply to
krw

Oh, that doesn't look too bad. I don't like the straight ones. Just because a screwdriver sorta needs to be straight to work doesn't mean it's the best shape for a powered one.

Depends on where the smarts are. Li-Ion is picky.

A web search turns up a ton of 9.6V Bosch batteries. I've found that some of the after market batteries are as good as or better than the originals. Though sometimes it's not worth replacing batteries. I've found deals on drills that are cheaper than the included batteries.

Reply to
krw

I find this one really useful for small stuff like building PCs.. the clutch works very nicely:-

formatting link

(looks like a good price-- I paid $14 more)

Best regards, Spehro Pefhany

--
"it's the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
 Click to see the full signature
Reply to
Spehro Pefhany

formatting link

recommend

formatting link

It matches these ones:

formatting link

(don't run with a reciprocating saw!)

Best regards, Spehro Pefhany

--
"it's the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
 Click to see the full signature
Reply to
Spehro Pefhany

I use PING (Partimage Is Not Ghost), which is a small Linux live disc that does incremental backups onto one of my small NAS devices. That's quite slow compared to even a USB HDD, but I have a cron script that syncs all the NASes up every night, which is comforting.

Cheers,

Phil Hobbs

Reply to
Phil Hobbs

My recollection is that Makita started all this with their kinda drab blue, Dewalt responded with bright yellow, leaving Hitachi to have to really go crazy with bright green and racing stripes. :-)

My least favorite tool color is probably Harbor Freight orange though! Some of their newer tools have switched to more reasonable silver/grey color schemes.

Reply to
Joel Koltner

You should see Harborfreight red. Yecch. Bought half a dozen meters. In the ad they were yellow, but "Chinese red" ones (same specs otherwise) came :-(

I hope that it at least reduces the theft likelihood, probably nobody wants to have that color.

--
Regards, Joerg

http://www.analogconsultants.com/
 Click to see the full signature
Reply to
Joerg
[...]

Exactamente. Got the same thing coming up for the digital cordless phones here, batteries dying again. The stores slap huge profit margins on those. When they want $15-$20 per battery and a 5-pack of new phones retails for around $100, well ...

--
Regards, Joerg

http://www.analogconsultants.com/
 Click to see the full signature
Reply to
Joerg
[...]

That's quite nifty and being slow would not bother me a bit. However, I don't like automatic processes for backup. Seen it happen too often that automatic systems copied from media that were already in the process of failure. The worst case was an engineer who had to re-do a large layout because the copy as well as the copy of the copy came up with "currupt data" when loading.

--
Regards, Joerg

http://www.analogconsultants.com/
 Click to see the full signature
Reply to
Joerg

Rigid orange isn't very appealing either.

"Harbor Freight" reduces the likelihood of theft. That's why I put them in my toolboxes. The Flukes stay hidden.

Reply to
krw

Just curious, incremental backups to what (image?) with what tool exactly? I usually do image backups of my Windows laptop with ntfsclone and then crunch some recovery data for the image with par2. But are you saying Partimage can do incremental NTFS images? Would save some time and space...

Reply to
Anssi Saari

There are several things to think about for a good backup strategy.

One is that it *should* be automatic - manual backups get forgotten, especially if you have to do something physical (like put a new tape in the machine).

Secondly, it should be properly monitored. Typically that means receiving an automated email after each backup with an unobtrusive "backup successful" message on success, or a more detailed message on failure.

Thirdly, you need a system that keeps lots of /logical/ copies (not necessarily lots of /physical/ copies). Ideally, you have lots more recent copies, and fewer copies over longer times (such as a set of tapes for each day of the week, then a set for once per week for a dozen weeks, then for once per month, etc.)

Fourthly, you need at least two totally independent physical copies, in different locations (covering hardware failure, fire, etc.)

Fifthly, it should be fast enough not to be a hinder to normal usage. Typically, that means running overnight - but watch out for issues like people working late, large changes leading to long delays, etc.

Finally, and most important of all, a backup strategy is only as good as the restore strategy. If the only method of restoring a file is a total file system restore from tape, or if it involves feeding in ten DVDs with incremental backups, it's useless.

Remember that the huge majority of data losses are due to human error, not hardware failure (or even software failure such as corrupt file systems). Data is lost because someone deletes the wrong file or folder on the server, and they then need yesterday's copy quickly and easily.

For many years I've been using rsync-based backup systems - previously dirvish, now rsnapshot. The system requires a *nix machine for the backup storage (since Windows does not support hard links).

Roughly speaking, the backup machine first takes a rsync copy of the server. Rsync only transfers over the changes to the file system, and for large files that have been modified, it only transfers the changes. Data is compressed and packed, making it very network-friendly.

Once that snapshot is complete (this takes about 10 or 15 minutes normally on my company network), a copy is made using "cp -al". This makes a copy using hard links - the entire directory tree is copied, but each file is hard-linked to the file in the snapshot, and thus the copy only takes up as much space as the file names and directory entries. The copied tree may then be renamed or moved depending on the details of the backup system (dirvish names the trees by date, rsnapshot calls them daily.0, daily.1, etc. for today's copy, yesterday's copy, etc.).

The point of this is that you can have dozens of snapshots for each day, week, month, year and it only takes as the files themselves plus any changes - it's as economical as always doing incremental backups. And yet each of these snapshot directories is a direct image of all your files - you can find exactly the files you want from the date you want using exactly the same tools as any other file system access. It's all instantly available on hard disk.

To complete the system, I've got two such backup servers - one at the office, and one at home doing backups overnight over the Internet.

Reply to
David Brown
[ much good advice about backup methods ]

...

Ideally the monitoring should include a check that files *can* be restored, or at least read from the backup. I have been hit once by a backup tape drive that seemed ok when writing tapes, but created unreadable tapes. Bummer to find that out when you need to restore stuff.

David's method of backup to disk memory, not tape, is of course much less likely to suffer from such hidden hardware problems.

--
Niklas Holsti
Tidorum Ltd
 Click to see the full signature
Reply to
Niklas Holsti

That's the point of the final and most important rule - your backup is only as good as your restores. If you are backing up to something inherently unreliable such as tapes or CDs/DVDs, you must regularly check that they are readable and that the backup has been successful. In particular, you must do the check on an independent machine, especially for tapes. If your tape drive's head alignment is off, the drive may appear to write fine, and read fine when checking the tapes - but when that machine and drive are damaged in a fire and you try to re-read on your replacement tape drive, it will fail.

To be extra careful, you can use mirrored raids on the backup disks if you want, but you should already have redundancy by having a second backup machine off-site. There is little need for double-redundancy - you'd have to be *very* unfortunate to get one site burgled and a disk crash on the other site at the same time.

Reply to
David Brown

Putting media in drives, gets forgotten in lots of places, due to only one member of staff does this who is then off sick/vacation. Other reasons include not a daily task, irregular task.

Classic example of irregular task is the BAD method some backup strategies use of whatever type, 'backing up' until a media is full then requests new media (tape/DVD..) and this gets missed. Let alone this process may take a week or two and during that time there is a fire/theft/flood/hardware failure.

Monitored preferably by more than one person, if your staff numbers are greater than one.

Don't keep all these copies on ONE media, too often I see ONE USB HDD used for all backup copies, when run overnight or just left by system being backed up unit is often plugged in. So the backup media is subject to

Electrical failures (spikes/lightning) that affect the system Fire Flood (down/up/sideways) Theft Malicious damage (I have seen this happen in some organisations) Failure of media/copy

Always keep your media even if some of them on site AWAY from the system. Many organisations of ANY size have had systems and the backup tapes left on top of the system were stolen with the system.

Multiple copies also covers failure of a copy (tape/drive failure)

Cleaners unplugging system to power a vacuum cleaner.

Check the strategy for backup does not include keeping the ONLY catalog for what is on what media stored on the hard drive being restored after a drive crash. 'One touch backup' software is too often like this.

If you are able to maintain a proveable two copies of system as snapshots that can be restored as a base system, this will save a lot of time for hard drive failures, I do about 4 a year for my small number of customers. This is often easier and quicker than finding all the discs for software to install or even the basic software to get a system configured to retsore latest full backups.

Determine your backup strategy (which includes restore strategy), based on what is the maximum data loss you can withstand. In other words the cost in time and money between backup times, and how long it takes to get going again.

Make sure you have the necessary local copies of the software to restore backups onto replacement systems/hard drives.

I am reminded of years ago someone who always recited

"It's on the backup [media]"

When asked HOW they were going to restore onto blank drive.

Above all else TEST your restore strategy regularly, even if this going to a different system and restoring a selection of files from whatever media to temporary locations.

--
Paul Carpenter          | paul@pcserviceselectronics.co.uk
    PC Services
 Click to see the full signature
Reply to
Paul Carpenter

Been there, done that - the rules I've given here are often based on experience!

Agreed.

See the next rule...

If you are concerned about the theft of the data (rather than just the hardware), or have data that must have limited access (i.e., the IT folk should not see it), then you must of course consider encryption and the like.

Hopefully that's just a myth! But an UPS is a good idea anyway...

That always makes sense - you have to base your strategy on the actual needs. But the ability to quickly restore an accidentally deleted file is often extremely useful, yet seldom specified. For example, it might be specified that you take a backup of all data at least once a week, which is a reasonable requirement for disaster recovery. Quickly and easily recovering someone's lost data can make a big difference when needed - but specifying that you've got to have a backup strategy that allows restoring data that is maximum one day old, restored within 10 minutes, would normally be unreasonable.

Normally it is your data that must be protected - software can mostly be reinstalled or downloaded again. However, sometimes installation procedures are difficult, or the software is hard to get - obviously you must then also backup the installation files and/or the installed files.

Reply to
David Brown

....

Nope, over the years I have seen this (or similar events) three times. In one small office the plug for the server for 6 staff, was marked "Do not Unplug", however this actually was plugged into a trailing extension socket and no one thought to also put a label on the plug of the trailing extension socket.

They also had another incident there where the server would turn off randomly which transpired to be damaged UPS partly from some of the silly ways things had been wired in the building, main neutral point had not been secured down properly.

--
Paul Carpenter          | paul@pcserviceselectronics.co.uk
    PC Services
 Click to see the full signature
Reply to
Paul Carpenter

That's what checklists are for. In the same way that a pilot goes through the landing checklist that includes lowering the gear (that is usually not automated). Or in the same way we switch our fire alarm batteries here. Haven't forgotten that, ever, done it 24 times since we live here, religiously.

[...]
--
Regards, Joerg

http://www.analogconsultants.com/
 Click to see the full signature
Reply to
Joerg

Checklists are good for things that must be done manually, and must not be forgotten. Automation is good for things that can be done automatically. There is no point in doing things manually just for the sake of it if it can be done just as well automatically.

It's a different matter if it can't be done automatically, or the automation doesn't have appropriate monitoring, or if it is safety-critical (that's why aircraft routines *are* automated, and require manual checking as well, or they are manual, and have automated checks).

Reply to
David Brown

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.