On Thu, 5 Aug 2021, Dr. Nikolaus Klepp wrote: I installed a Samsung 870 EVO 1TB in a laptop with a AMD chipset. I learned that Samsung EVO SSDs and AMD chipsets do not play well together. I had to disable NCQ in order to get rid of error messages that indicated the potential for data loss. Samsung (EVO) + Intel chipsets should be fine though. More reading here: https://bugzilla.kernel.org/show_bug.cgi?id=201693 Gianluca ----------------------------------------------------- Gianluca Interlandi, PhD gianluca(a)u.washington.edu +1 (206) 685 4435 http://gianluca.today/ Department of Bioengineering University of Washington, Seattle WA U.S.A. -----------------------------------------------------

Borg Labs wrote: I recommend only WD RED (if possible), but pay attention anyway what you are buying. I had Samsung and Seagate before, but all failed except of one. I replaced them with WD RED over time and these are working since without issues. I must say respect to WD. This is why I bought last year a couple of WD RED SSDs 1TB and replaced the spinning one. I am very happy with the setup. It serves the whole family very well since around 2005 ... but it underwent a serious evolution until it took a shape. In fact I bought two LSI controllers and 2x6 cages. I have only 8 disks inside and they are mirrored so that each half is on separate controller. LSI limits the speed to 3Gbps, but I feel better knowing that only if the mainboard or power supply fails, I would have an issue. I was wondering if I should buy a second mainboard, but after so many years, I would not mind to replace it if it fails.

Anno domini 2021 Thu, 5 Aug 14:03:01 -0400 Borg Labs scripsit: Please read comment #69. Actually I have a MS-7C56/B550-A PRO with Ryzen 5 3600 and the said Samsung 870 QVO 1TB - but YMMV :) Nik -- Please do not email me anything that you are not comfortable also sharing with the NSA, CIA ...

said Gianluca Interlandi: | I installed a Samsung 870 EVO 1TB in a laptop with a AMD chipset. I | learned that Samsung EVO SSDs and AMD chipsets do not play well | together. I had to disable NCQ in order to get rid of error messages | that indicated the potential for data loss. Samsung (EVO) + Intel | chipsets should be fine though. More reading here: | | https://bugzilla.kernel.org/show_bug.cgi?id=201693 Good to know, in that I'm running an AMD chipset on the machine in question. Does this extent to other makers, or is it just a Samsung thing, do you know? Wonder if the problem has gotten fixed in later kernels; the report is from 2018. Another question: I'm running ext4 exclusively here, but I wonder: is there a particular advantage in one filesystem over another on an SSD? -- dep Pictures: http://www.ipernity.com/doc/depscribe/album Column: https://ofb.biz/author/dep/

Anno domini 2021 Thu, 05 Aug 18:09:10 +0000 dep scripsit: just use "noatime" to reduce writes and move /tmp to tmpfs (as has already been stated). Nik -- Please do not email me anything that you are not comfortable also sharing with the NSA, CIA ...

On 8/5/21 10:30 AM, Gianluca Interlandi wrote: I've had only one compatibility problem with SSDs. It was an Inland Professional SSD on an AMD-based HP Probook laptop. A Silicon Power SSD worked fine. -- PGP key: http://homestead-products.com/pubkey.htm

On Fri, 6 Aug 2021, Thierry de Coulon wrote: You are right. It should have been "some (older) AMD chipsets". Gianluca ----------------------------------------------------- Gianluca Interlandi, PhD gianluca(a)u.washington.edu +1 (206) 685 4435 http://gianluca.today/ Department of Bioengineering University of Washington, Seattle WA U.S.A. -----------------------------------------------------

On Fri, 6 Aug 2021, dep wrote: I did send a link that you could have used for further reading: https://bugzilla.kernel.org/show_bug.cgi?id=201693 I was just sharing my personal experience to be helpful, but there is no way I can tell you whether you would have had no problems if you had gone the Samsung route. You can always send back what you ordered and get a Samsung, but you are the one who will have to deal with it if there is an incompatibility. If you want to know the answer for sure you would have to contact Samsung and AMD. I don't think anyone on this list has ever tested all possible combinations to tell what works and what doesn't. Gianluca ----------------------------------------------------- Gianluca Interlandi, PhD gianluca(a)u.washington.edu +1 (206) 685 4435 http://gianluca.today/ Department of Bioengineering University of Washington, Seattle WA U.S.A. -----------------------------------------------------

Hi dep, Sorry, I had misunderstood. I totally share your feelings. Samsung did not do the legwork to figure out which chipsets their controller works with. In the openSUSE thread I got the recommendation that phison controllers work well with AMD chipsets, but I have a hard time finding SSDs with a phison controller. Gianluca On Fri, 6 Aug 2021, dep wrote: ----------------------------------------------------- Gianluca Interlandi, PhD gianluca(a)u.washington.edu +1 (206) 685 4435 http://gianluca.today/ Department of Bioengineering University of Washington, Seattle WA U.S.A. -----------------------------------------------------

On Thursday 05 August 2021 01:02:54 pm Dr. Nikolaus Klepp wrote: Agreed for long term backups, use a mechanical drive. SSDs are too you to know how long term disuse will affect them. Kate

dep wrote: I wouldn't care that much, because these are robust devices and can handle 100s of 1000s writes. Note that WD BLUE != WD RED. WD RED is more expensive but apparently much better. But the one of the 500GB SSDs is WD BLUE and no issues so far. The 120GB is WD GREEN and also no issues although I do not keep this machine running all the time. It depends how you setup grub on both disks ... If I were you, I would clone the one to the other. Actually if you want to put only / and swap on the disk, you could use much smaller device - like 120GB or even 50GB would be more than enough. You could buy two smaller disks for the money for 1x500GB and RAID1 and forget about copies. Still make backups at least before major upgrades. regards PS: Here is smartctl output from the WD BLUE (and don't ask why it is 3Gbps) smartctl -a /dev/sda smartctl 6.6 2017-11-05 r4594 [x86_64-linux-4.19.190eko4] (local build) Copyright (C) 2002-17, Bruce Allen, Christian Franke, www.smartmontools.org === START OF INFORMATION SECTION === Model Family: WD Blue PC SSD Device Model: WDC WDS500G1B0A-00H9H0 Serial Number: xxxx LU WWN Device Id: 5 001b44 8b49c2a69 Firmware Version: X41100WD User Capacity: 500,107,862,016 bytes [500 GB] Sector Size: 512 bytes logical/physical Rotation Rate: Solid State Device Form Factor: 2.5 inches Device is: In smartctl database [for details use: -P show] ATA Version is: ACS-2 T13/2015-D revision 3 SATA Version is: SATA 3.2, 6.0 Gb/s (current: 3.0 Gb/s) Local Time is: Thu Aug 5 20:18:45 2021 CEST SMART support is: Available - device has SMART capability. SMART support is: Enabled === START OF READ SMART DATA SECTION === SMART overall-health self-assessment test result: PASSED General SMART Values: Offline data collection status: (0x00) Offline data collection activity was never started. Auto Offline Data Collection: Disabled. Self-test execution status: ( 0) The previous self-test routine completed without error or no self-test has ever been run. Total time to complete Offline data collection: ( 0) seconds. Offline data collection capabilities: (0x11) SMART execute Offline immediate. No Auto Offline data collection support. Suspend Offline collection upon new command. No Offline surface scan supported. Self-test supported. No Conveyance Self-test supported. No Selective Self-test supported. SMART capabilities: (0x0003) Saves SMART data before entering power-saving mode. Supports SMART auto save timer. Error logging capability: (0x01) Error logging supported. General Purpose Logging supported. Short self-test routine recommended polling time: ( 2) minutes. Extended self-test routine recommended polling time: ( 10) minutes. SMART Attributes Data Structure revision number: 4 Vendor Specific SMART Attributes with Thresholds: ID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE UPDATED WHEN_FAILED RAW_VALUE 5 Reallocated_Sector_Ct 0x0032 100 100 --- Old_age Always - 0 9 Power_On_Hours 0x0032 100 100 --- Old_age Always - 33740 12 Power_Cycle_Count 0x0032 100 100 --- Old_age Always - 653 165 Block_Erase_Count 0x0032 100 100 --- Old_age Always - 47120952 166 Minimum_PE_Cycles_TLC 0x0032 100 100 --- Old_age Always - 2 167 Max_Bad_Blocks_per_Die 0x0032 100 100 --- Old_age Always - 30 168 Maximum_PE_Cycles_TLC 0x0032 100 100 --- Old_age Always - 22 169 Total_Bad_Blocks 0x0032 100 100 --- Old_age Always - 517 170 Grown_Bad_Blocks 0x0032 100 100 --- Old_age Always - 0 171 Program_Fail_Count 0x0032 100 100 --- Old_age Always - 0 172 Erase_Fail_Count 0x0032 100 100 --- Old_age Always - 0 173 Average_PE_Cycles_TLC 0x0032 100 100 --- Old_age Always - 6 174 Unexpected_Power_Loss 0x0032 100 100 --- Old_age Always - 65 184 End-to-End_Error 0x0032 100 100 --- Old_age Always - 0 187 Reported_Uncorrect 0x0032 100 100 --- Old_age Always - 0 188 Command_Timeout 0x0032 100 100 --- Old_age Always - 0 194 Temperature_Celsius 0x0022 058 054 --- Old_age Always - 42 (Min/Max 17/54) 199 UDMA_CRC_Error_Count 0x0032 100 100 --- Old_age Always - 0 230 Media_Wearout_Indicator 0x0032 100 100 --- Old_age Always - 0x013e0055013e 232 Available_Reservd_Space 0x0033 100 100 004 Pre-fail Always - 100 233 NAND_GB_Written_TLC 0x0032 100 100 --- Old_age Always - 3467 234 NAND_GB_Written_SLC 0x0032 100 100 --- Old_age Always - 6181 241 Total_Host_GB_Written 0x0030 253 253 --- Old_age Offline - 5075 242 Total_Host_GB_Read 0x0030 253 253 --- Old_age Offline - 18988 244 Temp_Throttle_Status 0x0032 000 100 --- Old_age Always - 0 SMART Error Log Version: 1 No Errors Logged SMART Self-test log structure revision number 1 No self-tests have been logged. [To run self-tests, use: smartctl -t] Selective Self-tests/Logging not supported

On Thu, 5 Aug 2021, dep wrote: I know that at least for the mechanical spinning drives WD Reds are not really recommended for desktop usage (only for NAS or storage). For example, they have something called TLER (that can be disabled). But not sure about the SSDs. I would do some research and maybe even contact Western Digital? Not saying you shoudn't, but I would read about it a bit first. Gianluca ----------------------------------------------------- Gianluca Interlandi, PhD gianluca(a)u.washington.edu +1 (206) 685 4435 http://gianluca.today/ Department of Bioengineering University of Washington, Seattle WA U.S.A. -----------------------------------------------------

On Thursday 05 August 2021 12:53:51 Dan Youngquist wrote: I used to swear by WD drives, especially their higher-grade stuff; and so far, I have never had any of them fail on me. When I bought an SSD, I passed over cheaper Samsung and other brands, and went for a WD green. Seagate has been much worse for me, as I've had two Seagates fail on me. Of course, this is over the course of about 30 years or so, and only two hard drives have failed, but they are both Seagates. But recently I read some stuff online about the dangers of connecting WD hard drives to the Internet. I am sure that others will know the details better than I, but the gist of it, what I recall, is that there is some kind of danger of losing all your data, ZAP! in the blink of an eye. This is due to some [I forget this part]. And the worst of it is that WD was removing support; although then I've heard that they are now offering support again for these devices or this problem. Since nobody else has mentioned it, I thought it's worth saying. I don't know if it applies to SSDs; I think it included only their spinning drives, but I could be wrong. If pressed for my sources, this will take some time, but I hope that others out there will know what I'm trying to relate, and can provide links for further reading. Bill

On Thursday 05 August 2021 16:00:15 Dan Youngquist wrote: Or even hang there, my 240G amanda holding disk was temporarily installed by hanging it on the /dev/sdc cables. And I'm a bit lazy, so about 9 months later its still hanging there. It did solve the problem of bad checksums in the holding disk, which was what I was watching amanda fight with, and which caused amanda to make a second pass at backing up that disklist entry, and leaving the failed one there gradually using up the holding disk space. And worse, it wrote both backups to spinning rust. The last such incident was in feb, so all such confusion has since been overwritten by my 60 tape vtape setup on a 2T spinning rust drive, a seagate of course. Cheers, Gene Heskett -- "There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." -Ed Howdershelt (Author) If we desire respect for the law, we must first make the law respectable. - Louis D. Brandeis Genes Web page <http://geneslinuxbox.net:6309/gene>

Well, I just ordered the WD Red, 500gb. Chose that one because of deloptes's recommendation and because best I can tell this is as sturdy as they get. (As, it seems, is attested in the warranty.) By the time it gets here I hope I'll have cooked up a good fstab for it -- the existing drives will have to move down by one letter, so my sda4 will become sdb4 b4 I can use it. And so on. I hope the increase in speed will be noticeable. Incremantal increases in computer speed in my experience tend to be inverse-obvious. New fast computer: I *think* it's a little faster. Returning to previous computer for something six months later: My gawd, this thing is *s-l-o-w*! You'll hear way too much about my installing it before I get it done and will regret having answered my original post, but thanks to everybody. -- dep Pictures: http://www.ipernity.com/doc/depscribe/album Column: https://ofb.biz/author/dep/

Anno domini 2021 Fri, 6 Aug 14:44:26 +0200 phiebie(a)drei.at scripsit: And the UUIDs will give you a hard time when you duplicate drives and boot with two identical UUIDs. e.g.: dd if=/sda of=/dev/sdb Remove /dev/sda - it's the old one, replace it by /dev/sdb. Put the old drive on the shelfe for backup. Some time later you need data from the old drive (will never happen, for sure) and connect it to your system with a sata-usb-adapter or plug it into a free sata port of your mainboard. Reboot or mount the drive ... fun guaranteed :) That said, uuids are kind of coll, but they get in my way all the time, just like the "new" aka GNOMEish naming schemes for e.g. eth0. Nik -- Please do not email me anything that you are not comfortable also sharing with the NSA, CIA ...

Anno domini 2021 Fri, 06 Aug 14:10:51 +0000 dep scripsit: It's just to make sure you do stuff on the right drive. UUIDs should be copied by clong software, too, otherwise grub of your clone will not boot. So you'll have to chane UUIDs of the clone, adjust grub and check that it's bootable. dd works like a charm, as long as your target drive is at least of the same size as the source drive. You can do some magic with gpart, e.g. shrink partitions etc., but you'd need to start with something like puppy or exegnulinux or anything else with gpart on it. Nik -- Please do not email me anything that you are not comfortable also sharing with the NSA, CIA ...

On Friday 06 August 2021 07:10:51 dep wrote: regarding cloning (partitions, or the whole drive): I have an older version of this nifty little gadget, and am considering buying one of the new models. It works very well, and the newer models have a "one-touch cloning" feature. The only problem I ever had was that the hard drives themselves tend to overheat, especially in warm weather (because they are not inside the chassis with a fan blowing on them). Sometimes I would even take the overheated hard drive and set it inside my freezer for a little while to cool down, because they would get HOT to the touch. Solution: get a powerful fan, and set your docking station directly in front of it. However, you don't want to use this full-time, but only for plugging in and transferring files, backups or cloning a partition, etc. You shouldn't, for example, try to watch films from a hard drive plugged into your docking staton. Bill

On Friday 06 August 2021 11:08:37 dep wrote: I never had one of those hard drives fail on me, but then I didn't keep using it like that (watching films, etc.). After that, I used the docking station for its intended purpose, not to make instant external hard drives. The Kingwin EZ Dock is (for me, at least) definitely a keeper, and does file transfers, backups and cloning very well. So I thought maybe it could be useful to you. They are only $35 or so, last I checked. But, like I said, keep a fan directed right on it, and don't use it like external hard drives. Bill

On Friday 06 August 2021 11:40:33 dep wrote: OS/2 Now there's a blast from the past. Bill

On Friday 06 August 2021 07:10:51 dep wrote: regarding cloning (partitions, or the whole drive): I have an older version of this nifty little gadget, and am considering buying one of the new models. [Sorry, forget to add this link!] http://www.kingwin.com/storage/docking-stations/ezd-2535u3-2/ It works very well, and the newer models have a "one-touch cloning" feature. The only problem I ever had was that the hard drives themselves tend to overheat, especially in warm weather (because they are not inside the chassis with a fan blowing on them). Sometimes I would even take the overheated hard drive and set it inside my freezer for a little while to cool down, because they would get HOT to the touch. Solution: get a powerful fan, and set your docking station directly in front of it. However, you don't want to use this full-time, but only for plugging in and transferring files, backups or cloning a partition, etc. You shouldn't, for example, try to watch films from a hard drive plugged into your docking staton. Bill

said Gene Heskett: | I much prefer to use the LABEL=name, with a unique name for that | partition, it seems immune to everything but a partition table erasing | full format. This is useful in fstab? GRUB? Here are a couple of entries from my fstab: # <file system> <mount point> <type> <options> <dump> <pass> /dev/sda1 / ext4 errors=remount-ro 0 1 UUID=8C5E-D456 /boot/efi vfat defaults 0 1 I would replace the <file system> value with LABEL=[name i gave the partition] and GRUB and everybody would be happy with that? If not, what would the syntax be to make LABEL= useful? -- dep Pictures: http://www.ipernity.com/doc/depscribe/album Column: https://ofb.biz/author/dep/

On Thu, Aug 05, 2021 at 04:07:51PM -0400, Gene Heskett wrote: Do you think that they are using the same corrosive dye in 2021 in SATA cables that people used in CB radios in 1970-something? I don't know what you consider "hot red" versus some other red, but I've had a PC with red SATA cables work fine for six or seven years until an unrelated fault forced me to retire the machine. -- Steve

Gene Heskett wrote: whos the manufacturer?

dep wrote: if it is recent, it should work - I know what Gene will say, but I do not think nowdays there is a problem with the cables. I forgot to mention another advantage of SSD - the power consumption.

said phiebie(a)drei.at: | SSD's have a wear-levelling algorithm. That means, very very short | explained, *all* cells get written to the same amount of times and | therefore wear out evenly. Put swap on it. | But to extend the lifespan | of the SSD, don't partition the total 500GB, stop at let's say 475GB. | These 25GB can then be used by the fault-detection of the SSD to | compensate for "rotten" cells in the workspace. So, then, its spare sectors come from unpartitioned, unformatted space? | Put everything, that is executable or needed by other programs on the | SSD! Only the pure data (photos aso.) should stay on the harddrives. My plan is to put everything except /home there. I'm a little puzzled as to the partitioning map, in part because of my switch to a 6gb boot drive earlier this year, which required creation of an efi partition (which I still don't much understand). I'm hoping to create a single partition, mount it as / , and call it a day, with /home and other stuff on traditional drives. I further hope to keep my existing boot drive, but not boot from it. In the happy days of yore, it would show up on the GRUB menu at boot, so if I wanted I could boot from it. But since the switch to the bigger boot drive and upgrade to 20.04-LTS I no longer have a grub menu. Might this be because I no longer have a dual-boot system? (I nuked the XP partition because I hadn't booted to it in years.) | Maybe you find another, better solution, but I would suggest: | Make one harddrive bootable | Install on the SSD backintime with as backup-medium that harddrive | After having apt-upgraded run backintime and the harddrive should be a | mirror of the SSD. Ah. Makes sense. This would work, then, with my existing boot drive to make a good fallback, presuming I can recover my GRUB menu, no? | When you're about adjusting the fstab: don't use the old driveletters | anymore, a pita sometimes, take partition-UUID's, they stay the same | whatever you shuffle drives around. Also makes sense -- means among other things that I wouldn't need a different one for the bootable backup. Also obviates another question I had (but figured I'd figure out), which is whether a drive's physical location in the SATA system mattered in assigning drive designations. If I read you accurately, by using UUIDs this wouldn't matter (if it did to begin with -- I wasn't paying a lot of attention when we switched from IDE to SATA, and since then all I've done is replace existing drives with bigger drives). -- dep Pictures: http://www.ipernity.com/doc/depscribe/album Column: https://ofb.biz/author/dep/

said phiebie(a)drei.at: | Run update-grub. I have, repeatedly. Last line of the result is: "Adding boot menu entry for UEFI Firmware Settings". Problem is, the GRUB menu never displays. Doesn't show up at boot time. I am not sure how to make it do so -- until now, there was no reason to mourn its passing. I knew what to do with LILO, but I try not to screw around much with GRUB, because "much" can become "too much" pretty easily. | Nik mentioned a case, where this wouldn't work as expected. But | Linux prefers to use UUID (Universally Unique Identifier), LABEL, or | symlinks to identify media storage devices on a system. Using device | assignments (like /dev/hd*# or /dev/sd*#) is not preferred since these | can change between system boots. Right, and I understand that. But there still are puzzles, such as the partition ID, which I suspect is useful but have now idea how. -- dep Pictures: http://www.ipernity.com/doc/depscribe/album Column: https://ofb.biz/author/dep/

On Fri, 6 Aug 2021 19:25:45 +0200 "Dr. Nikolaus Klepp" &lt;office(a)klepp.biz&gt; wrote: Externals can change their device nodes randomly (I have a flaky Pi + enclosure combination where the drive can show up as sda or sdb depending on what it ate for breakfast). I suppose there are some people for whom that matters, and I do use a label in the autofs configuration on that flaky Pi (the only place I've ever used one). I've never seen an internal drive change its device node without a hardware change, but I suppose it would in theory be possible for it to happen if the system for some reason initialized an external drive attachment point first, or a drive ahead of it in the initialization order failed so hard that the system could no longer detect it. Regardless, I would class it as highly unusual on a system that doesn't have hot-swap drive bays or the like. Gentoo documentation doesn't indicate a preference for one type of drive identifier over the others. We could perform a survey, I suppose. E. Liddell

Hi Nik, They DID and DO! Right now still. I have Debian stable, without any tinkering anywhere in the whole system after installation, on my third harddrive, being sdc. 3 out of 10 times when booting into stable I get between the first few screen-messages "/dev/sdb2 clean" and "/dev/sdb5 clean". B not C. Neglecting the wrong letters and starting the system brings me where I wanted to be, into stable. Reason: in the grub.cfg there is no mentioning of /dev/sdx, only hdx and the relevant UUID's so the system knows where to look. No myth therefore. Regards Peter.

Borg Labs wrote: ahm ... 1TB for / ... sorry but no, it doesn't make sense.

Anno domini 2021 Thu, 5 Aug 14:44:57 -0400 Gene Heskett scripsit: Soft-raid? Better make /boot first partition of equal size on each drive & make it a mirror, so you can boot when a drive fails. nik -- Please do not email me anything that you are not comfortable also sharing with the NSA, CIA ...

On 8/5/21 11:59 PM, Felix Miata wrote: It's worth noting that all those you've had fail are brands that don't have very good reputations in SSDs. Stay away from those and you'll have a much better failure rate. -- PGP key: http://homestead-products.com/pubkey.htm

Anno domini 2021 Fri, 6 Aug 13:21:59 -0400 Gene Heskett scripsit: Keep an eye on that 60GB if does not have SMART. These old devices die without warning. When they do, you suddenly have a 16MB unformated drive insted of 60GB - that's for flashing the firmware to recover your drive :) Nik -- Please do not email me anything that you are not comfortable also sharing with the NSA, CIA ...

said Gene Heskett: | All are working flawlessly after I purged my | systems of sata-usb3 cables of other brands in favor of startech's. Maybe you could remove that ozone-generator card from your system and your cables wouldn't be so sensitive.<g> -- dep Pictures: http://www.ipernity.com/doc/depscribe/album Column: https://ofb.biz/author/dep/

said Gene Heskett: | On Friday 06 August 2021 14:16:49 dep wrote: | > said Gene Heskett: | > | All are working flawlessly after I purged my | > | systems of sata-usb3 cables of other brands in favor of startech's. | > | > Maybe you could remove that ozone-generator card from your system and | > your cables wouldn't be so sensitive.<g> | > -- | > dep | | As one smartass to another, gladly if you can identify which one it | is. ;o) What? You don't have a dedicated ozone generator card? If you're relying on one of the all-in-ones that produce ozone, electrical noise, imtermittent power drops, and a high-pitched whine, well, there's your problem. Those things are supposed to be generated separately. -- dep Pictures: http://www.ipernity.com/doc/depscribe/album Column: https://ofb.biz/author/dep/

On Saturday 07 August 2021 11:14:49 dep wrote: Well, pretty much all of that is all-in-one here, with a Generac label on it, good for 20 kw running on nat gas, out in the back yard, with a 1500 WA ups insulating this box from such powerline shennanigans. It was needed several years ago by the wife's oxygen generator as she had COPD, finally losing that battle last Dec 7th, Pearl Harbor Day, which I as an old fart, can remember from 1941. It was the first and last time I ever saw my grandfather cry as we listened to the 6 oclock news on a battery powered radio on a farm 10 miles from any commercial electricity. He knew we were going to war, but at 7 yo, I didn't really know what "war" was. But I sure found out. Moral of this story is that you shouldn't try to out smartass a genuine Senior Citizen, he just might out do you by telling the truth. OTOH, at 86, I am used to being the Senior member of any mailing list I am subscribed to. About 40 of them. That, and activities normally attributed to people half my age, all connive to keep me out of the bars. Which is a Good Thing since we no longer have pills for some of the stuff you can get after hours. Take care now. Cheers, Gene Heskett -- "There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." -Ed Howdershelt (Author) If we desire respect for the law, we must first make the law respectable. - Louis D. Brandeis Genes Web page <http://geneslinuxbox.net:6309/gene>