Re: sata driver compataility Q

[gene heskett]

On 9/18/23 07:57, Dan Ritter wrote:

gene heskett wrote:

[...]

It looks like the motherboard shares some PCIe and/or SATA lanes between
SATA ports and M.2 ports, so you must be careful with your choices.?? I
suggest installing an M.2 PCIe x4 SSD into slot M.2_1 and configuring it
for "PCIE mode", so that it works and all 6 SATA ports work.?? You will
want to use EUFI mode and GPT when installing Debian.



Based on this, and a full sized manual printout,  I've ordered a 2T WD
Black, supposedly a 2280 device. $100.

Question, when I put this in, what happens to the 32GB of dimms? How does
this fit into the architecture?  I assume this isn't volatile but is quick
storage.


The DIMM slots are different from the M.2 slots. The M.2 slots
are small PCIe interfaces; the installation procedure is to
insert the 2280 (22mm x 80mm) card at a slight upward angle,
then press it down and screw it in. It may ship with a glued-on
heat spreader or tiny radiator; if so, use it, don't peel it
off.

Note that it should appear as /dev/nvme0n1 or similar, rather
than /dev/sda. Partitions will be /dev/nvme0n1p1, p2...

The NVM bits stands for non-volatile memory; it's an SSD with a
different interface.


I propose to put this in as suggested, which should leave all 6 sata-III's
available, Install bookworm to it, w/o the current ectra controller get it
going, then put 3 of the 2T gigastones on sata1-2-3, use the bios to make a
raid5 of them and mount it as /home, prove it works with some throw away
stuff, then plug the existing raid10 controller & mount it as moi, then
format the raid5 again with gparted,


You should use mdadm rather than a BIOS RAID system -- better
recovery to other systems, more understandable error messages,
better support for fixing things that might go wrong.


Thanks, I will.


If the new drives are sda through sdf something like this is what
you want:

mdadm --create /dev/md/gsmoi5 -l raid5 -n 3 /dev/sda /dev/sdb /dev/sdc
mkfs.ext4 /dev/md/gsmoi5
mount /dev/md/gsmoi5 /home


and for your second set:

mdadm --create /dev/md/gsamanda5 -l raid5 -n 3 /dev/sdd /dev/sde /dev/sdf
mkfs.ext4 /dev/md/gsamanda5
mount /dev/md/gsamanda5 /amanda



Suggestions re other, more recent solutions will be accepted and studied.
Definitely must support backing up other machines of varying architectures
on my local network. In addition to a 4 pack of linux running wintel stuff,
there's the potential for 5 or so arms too. Gcodes for 3d printers are all
unrolled loops and bulky as can be.


The most flexible backup systems are the hardest to configure,
but nothing is much worse than amanda.

You might like borg. Borg is in Debian as 'borgbackup'.

In the other direction, using rsnapshot over ssh is relatively
simple and comes with the distinct advantage over both amanda
and borg that the backups are stored as normal files in a normal
filesystem, so recovery from an accidental deletion of a file or
directory is very straightforward.

Compared to the setup required for amanda, that sounds very inviting. 
Amanda has a very steep learning curve just because it is so versatile 
I'm still waiting on stuff, so no more actual progress.


Thanks Dan, tae care & stay well.

-dsr-
.


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, 1940)
If we desire respect for the law, we must first make the law respectable.
 - Louis D. Brandeis
Genes Web page 

Httpd base image, libhiredis version

[Xu, Fong]

Hi,

I am reaching out is to ask if you have any other versions of libhiredis that 
we can use.

We are having CVE with our httpd base image. When we upgrade to the latest 
version of httpd, the CVE is gone. However, our app will not start. The error 
we are getting is related to redis. From the list that we can find, 
libhiredis0.14 (0.14.1-4) is the only version we can see. So we had to use the 
earlier version - httpd:2.4.56-debian-ff which work with libhiredis0.14. Any 
suggestions?

Here is what we have currently:

apt install -y libcjose0 libapache2-mod-auth-openidc libhiredis0.14

https://packages.debian.org/sid/libs/

https://packages.debian.org/sid/libhiredis0.14


Please let me know if you need more information.

Thank you,
Fong

Re: memtest86

[Anssi Saari]

Stefan Monnier  writes:

> Of course the `.com` version won't push all its own caveats in your
> face, that would go against its own commercial interests.

Indeed. Like a feature it has might not work due to other reasons. For
example, I bought the memtest86 pro since I wanted to use the ECC error
injection feature to see ECC memory correction in action. Unfortunately,
on two of my ECC computers error injection isn't working (meaning it's
disabled and there's no option to enable) and the third is old school
and doesn't have UEFI so I can't boot memtest86 pro on it.

Not to mention the GUI seemed unusable on both the systems I tried so
basically I had to do batch testing. At least that limitation was
mentioned somewhere.

Re: resizing PDF files with Ghostscript

[Curt]

On 2023-09-17, Greg Marks  wrote:
>
> I am trying to use Ghostscript to resize PDF files to letter page size,
> but on certain files the output is not the correct size.  As an example:
>
>$wget https://gmarks.org/abrams_anh_pardo.pdf
>
>$pdfinfo abrams_anh_pardo.pdf=20
>...
>Page size:  539 x 737 pts
>...
>
>$gs -o resized_file.pdf -sDEVICE=pdfwrite -dFIXEDMEDIA -dPDFFitPage -d
> DEVICEWIDTHPOINTS=612 -dDEVICEHEIGHTPOINTS=792 -dBATCH -dSAFER abrams_a
> nh_pardo.pdf
>
>$pdfinfo resized_file.pdf=20
>...
>Page size:  579.224 x 792 pts
>...
>
 
It appears the output of pdfwrite has a MediaBox of 612x792 but a CropBox of
595.61x792, which must be the result of scaling A4 down to Letter and
centering that scaled down area on the page.

curty@einstein:~$ pdfinfo -box resized_file.pdf 

Page size:  579.224 x 792 pts
Page rot:   0
MediaBox:   0.00 0.00   612.00   792.00
CropBox:   16.39 0.00   595.61   792.00
BleedBox:  16.39 0.00   595.61   792.00
TrimBox:   16.39 0.00   595.61   792.00
ArtBox:16.39 0.00   595.61   792.00
File size:  26538 bytes
Optimized:  no
PDF version:1.7

I'm unaware of another way of "fixing" this but the following kludge (unless
you can tell your printing process to use MediaBox and ignore CropBox).

 sed -e "/CropBox/,/]/s#.# #g" resized_file.pdf 

curty@einstein:~$ pdfinfo -box resized_file.pdf 

Page size:  612 x 792 pts (letter)
Page rot:   0
MediaBox:   0.00 0.00   612.00   792.00
CropBox:0.00 0.00   612.00   792.00
BleedBox:   0.00 0.00   612.00   792.00
TrimBox:0.00 0.00   612.00   792.00
ArtBox: 0.00 0.00   612.00   792.00
File size:  26538 bytes
Optimized:  no
PDF version:1.7

AW: Debugging initramfs, server hangs during boot process

[thah...@t-online.de]

I have booted on 6.3 and 6.4, both with the same result.
If I boot with 6.1 the prompt for entering the luks passphrase is there and 
I can use the machine.
Getting 6.3 using snapshot.debian.org is possible, but also needs some changes 
on 
the packages config. I managed, but this needs additional work to get it right.


I have filed a bug against udev and later vs. nvidia-driver.
Purging nvidia stuff and using nouveau I could also run with 6.3 o 6.4.
After 6.5 came into debian I have tried again with nvidia and it works again.
So for 6.3 and 6.4 I basically couldn't use nvidia. Not sure about 6.2 as this 
is
not available on debian.

To diagnose the issue I always booted into debian live.
Mounted the encrypted luks partition and modified the initramfs-tools, generated
new initrd-Images and try again and again.
Getting early netconsole to work and fiddling with the initramfs stuff was not 
so easy for me.
I know that there can be issues running sid, but this was a lot harder than 
expected.
You have a working system, do an upgrade and boom this system cannot even boot 
up.

Using debug=x (or whatever value, it doesn't matter) and netconsole=... helped 
to remedy
this problem.

Sigh, Thomas
 


-Original-Nachricht-
Betreff: Re: AW: Debugging initramfs, server hangs during boot process
Datum: 2023-08-30T12:00:56+0200
Von: "Michel Verdier" 
An: "debian-user@lists.debian.org" 

On 2023-08-30, thah...@t-online.de wrote:

> The last USB device in this list is a bluetooth card. I have blacklisted 
> btusb, 
> but this didn't help, still hangs.
> At 182 seconds I pressed CRTL_ALT_DEL and the times without sleep statements 
> come down to like 11 seconds when it hangs.

You should blacklist usb devices not found during boot and still listed
with udevadm. You can also try to unplug all usb devices and plug them
one by one to find which hangs.
Do you try to boot on another kernel or a debian live ?



Re: Can't mount NFS NAS after major upgrade

[debian-user]

Steve Matzura  wrote:
 
> mount /mnt/bigvol1/dir-1 /home/steve/dir-1 -o bind,ro

In addition to what others have observed it might be worth mentioning
that the -v option to mount (i.e. verbose) often gives more information
about what's going on.

Re: sata driver compataility Q

[David Christensen]

On 9/17/23 18:17, gene heskett wrote:

On 9/17/23 17:52, David Christensen wrote:

On 9/17/23 03:26, gene heskett wrote:

On 9/16/23 19:46, David Christensen wrote:

On 9/15/23 19:37, gene heskett wrote:

On 9/15/23 20:12, David Christensen wrote:

On 9/15/23 15:04, gene heskett wrote:

On 9/15/23 17:35, David Christensen wrote:

On 9/15/23 12:28, gene heskett wrote:

I've just ordered some stuff to rebuild or expand my Raid setup.
This 16 port sata-III pci-e card:

along with a bigger drive cage, cables and such and some 
gigastone 2T drives to make a raid big enough to run amanda. 
And maybe put a new card in front of my 2T /home raid10.



... Asus PRIME Z370-A II motherboard ...



... i5-9600K CPU @ 3.70GHz ...


A fresh install of  Debian stable or old-stable should solve the 
storage I/O stuttering problems you are experiencing.


It looks like the motherboard shares some PCIe and/or SATA lanes 
between SATA ports and M.2 ports, so you must be careful with your 
choices.  I suggest installing an M.2 PCIe x4 SSD into slot M.2_1 and 
configuring it for "PCIE mode", so that it works and all 6 SATA ports 
work.  You will want to use EUFI mode and GPT when installing Debian.


Based on this, and a full sized manual printout,  I've ordered a 2T WD 
Black, supposedly a 2280 device. $100.



This one?

https://www.amazon.com/dp/B09QV5KJHV


Question, when I put this in, what happens to the 32GB of dimms?  How 
does this fit into the architecture?  I assume this isn't volatile but 
is quick storage.



You will still have 32GB of memory.  The WD Black is a fast SSD.  The 
crux will be configuring your motherboard firmware Setup program so that 
d-i can see the WD Black during installation and so that the new Debian 
installation can boot and run.



I propose to put this in as suggested, which should leave all 6 
sata-III's available, Install bookworm to it, w/o the current ectra 
controller get it going, then put 3 of the 2T gigastones on sata1-2-3, 
use the bios to make a raid5 of them and mount it as /home, prove it 
works with some throw away stuff, then plug the existing raid10 
controller & mount it as moi, then format the raid5 again with gparted,
and turn mc loose copying /moi to /home to get my working data back. 
Then 3 more 2T gigastones on the last 3 mobo sata ports, make another 
raid5 out of those mounted as amandatapes.. Unforch, I had a wrapper 
around amanda that it took me 5 years to fine tune but I've no idea if a 
backup copy exists anyplace it this midden heap. Amanda, as it exists, 
if you start a recovery to bare metal, can only restore to yesterday, my 
wrapper was a special deal that grabbed the database fom the just 
finished backup and put that into the vtape, uncompressed which if that 
was untared to the bare metal gave anmanda the data for recovery that 
would bring the system back to this mornings state. It also cleaned out 
the database of any links that referenced vtapes that were recycled and 
re-used.



Most hardware- and hybrid hardware/ software RAID solutions expect 
Windows -- e.g. the manufacturer provides a Windows bundle with device 
driver, CLI, GUI, etc..  Looking at the Asus PRIME Z370-A II Driver & 
Tools page, I see various Windows packages related to storage, but 
nothing for Linux.  Unless you can find suitable Debian packages, I 
would advise against motherboard RAID.



Debian supports software RAID via md, LVM, and btrfs.  I suggest that 
you use one of those.



ZFS is another possibility, but the learning curve is non-trivial.


So while I'm familiar with amanda, its been sold to an outfit that 
doesn't care, so its getting long in the tooth with only user support.


Suggestions re other, more recent solutions will be accepted and 
studied.  Definitely must support backing up other machines of varying 
architectures on my local network. In addition to a 4 pack of linux 
running wintel stuff, there's the potential for 5 or so arms too. Gcodes 
for 3d printers are all unrolled loops and bulky as can be.



I suggest starting with the WD Black and the new Debian installation.  A 
fresh install on a new device will simplify re-arranging the rest of 
your disks later.  The challenge will be deciding what data to put on it 
after Debian boot, swap, and root; and if and how to subdivide the space.




Thank you David, take care and stay well.


Likewise.  :-)


David