Gentoo Forums
Gentoo Forums
Gentoo Forums
Quick Search: in
Should partition sizes be powers of 2 ?
View unanswered posts
View posts from last 24 hours

Goto page Previous  1, 2, 3  Next  
Reply to topic    Gentoo Forums Forum Index Other Things Gentoo
View previous topic :: View next topic  
Author Message
Tony0945
Watchman
Watchman


Joined: 25 Jul 2006
Posts: 5127
Location: Illinois, USA

PostPosted: Fri Jan 15, 2021 7:16 pm    Post subject: Reply with quote

figueroa,

That 2011 BIOS is the latest available. Although seeing how the board is dual BIOS, maybe I could change the BIOS jumper.
I want to remove the sda drive as it is damaged and can't complete a smart control check. I suppose I could boot off of sdb and just replace sda with a similar drive.
I haven't thought of that before.
Back to top
View user's profile Send private message
NeddySeagoon
Administrator
Administrator


Joined: 05 Jul 2003
Posts: 54455
Location: 56N 3W

PostPosted: Fri Jan 15, 2021 10:42 pm    Post subject: Reply with quote

Tony0945,

Can't you boot from a floppy any more :)
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.
Back to top
View user's profile Send private message
Tony0945
Watchman
Watchman


Joined: 25 Jul 2006
Posts: 5127
Location: Illinois, USA

PostPosted: Fri Jan 15, 2021 11:58 pm    Post subject: Reply with quote

NeddySeagoon wrote:
Tony0945,

Can't you boot from a floppy any more :)

No my floppy drive bit the dust.
Back to top
View user's profile Send private message
Tony0945
Watchman
Watchman


Joined: 25 Jul 2006
Posts: 5127
Location: Illinois, USA

PostPosted: Wed Jan 20, 2021 11:05 pm    Post subject: Reply with quote

Got the drive assembled and into the computer. well, actually, the side is off and parts are strewn around, but i got it recognzing all the drives and in the right order.

I've started copying the partitions but something doesn't look right. I must have hit keys wrong?
Code:
# fdisk -l /dev/sdc
Disk /dev/sdc: 931.53 GiB, 1000204886016 bytes, 1953525168 sectors
Disk model: CT1000MX500SSD1
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 4096 bytes
I/O size (minimum/optimal): 4096 bytes / 4096 bytes
Disklabel type: dos
Disk identifier: 0xefbea596

Device     Boot      Start        End    Sectors   Size Id Type
/dev/sdc1             2048 1509951487 1509949440   720G 83 Linux
/dev/sdc2       1509951488 1644169215  134217728    64G 83 Linux
/dev/sdc3       1711278080 1953523711  242245632 115.5G 83 Linux
/dev/sdc4       1644169216 1711278079   67108864    32G 83 Linux

Partition table entries are not in disk order.
Not in order indeed! It looks like sdc3 and sdc4 are reversed. Should I wipe the disk and start over?

EDIT:
Code:

~ # e2label /dev/sdc
e2label: Bad magic number in super-block while trying to open /dev/sdc
/dev/sdc contains `DOS/MBR boot sector; partition 1 : ID=0x83, start-CHS (0x4,4,1), end-CHS (0x3ff,254,2), startsector 2048, 1509949440 sectors; partition 2 : ID=0x83, start-CHS (0x3ff,254,2), end-CHS (0x3ff,254,2), startsector 1509951488, 134217728 sectors; partition 3 : ID=0x83, start-CHS (0x3ff,254,2), end-CHS (0x3ff,254,2), startsector 1711278080, 242245632 sectors; partition 4 : ID=0x83, start-CHS (0x3ff,254,2), end-CHS (0x3ff,254,2), startsector 1644169216, 67108864 sectors' data
Back to top
View user's profile Send private message
Tony0945
Watchman
Watchman


Joined: 25 Jul 2006
Posts: 5127
Location: Illinois, USA

PostPosted: Wed Jan 20, 2021 11:27 pm    Post subject: Reply with quote

Also parted
Code:
(parted) p                                                               
Model: ATA CT1000MX500SSD1 (scsi)
Disk /dev/sdc: 1000GB
Sector size (logical/physical): 512B/4096B
Partition Table: msdos
Disk Flags:

Number  Start   End     Size    Type     File system  Flags
 1      1049kB  773GB   773GB   primary  ext4
 2      773GB   842GB   68.7GB  primary  ext4
 4      842GB   876GB   34.4GB  primary  ext4
 3      876GB   1000GB  124GB   primary  ext4
Aargh! The final partition should have been the odd-sized one , but it's the only even sized one and it's #3 instead of #4.
Back to top
View user's profile Send private message
Tony0945
Watchman
Watchman


Joined: 25 Jul 2006
Posts: 5127
Location: Illinois, USA

PostPosted: Thu Jan 21, 2021 12:29 am    Post subject: Reply with quote

Googled and used parted, x for advanced and f for fix. Looks right now.
Code:
~ $ fdisk -l /dev/sdc
Disk /dev/sdc: 931.53 GiB, 1000204886016 bytes, 1953525168 sectors
Disk model: CT1000MX500SSD1
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 4096 bytes
I/O size (minimum/optimal): 4096 bytes / 4096 bytes
Disklabel type: dos
Disk identifier: 0xefbea596

Device     Boot      Start        End    Sectors   Size Id Type
/dev/sdc1             2048 1509951487 1509949440   720G 83 Linux
/dev/sdc2       1509951488 1644169215  134217728    64G 83 Linux
/dev/sdc3       1644169216 1711278079   67108864    32G 83 Linux
/dev/sdc4       1711278080 1953523711  242245632 115.5G 83 Linux
I had better run fschk on them. then maybe wipe and recopy? or rsync -van and see if there is anything to be done?
Don't like this part:
Code:
 ~ $ e2label /dev/sdc
e2label: Bad magic number in super-block while trying to open /dev/sdc
/dev/sdc contains `DOS/MBR boot sector; partition 1 : ID=0x83, start-CHS (0x4,4,1), end-CHS (0x3ff,254,2), startsector 2048, 1509949440 sectors; partition 2 : ID=0x83, start-CHS (0x3ff,254,2), end-CHS (0x3ff,254,2), startsector 1509951488, 134217728 sectors; partition 3 : ID=0x83, start-CHS (0x3ff,254,2), end-CHS (0x3ff,254,2), startsector 1644169216, 67108864 sectors; partition 4 : ID=0x83, start-CHS (0x3ff,254,2), end-CHS (0x3ff,254,2), startsector 1711278080, 242245632 sectors' data


Code:
ony@Casti ~ $ e2label /dev/sdc1
MX500ROOT
tony@Casti ~ $ e2label /dev/sdc2
32BIT
tony@Casti ~ $ e2label /dev/sdc3
UBUNTU
tony@Casti ~ $ e2label /dev/sdc4
RESCUE
Back to top
View user's profile Send private message
NeddySeagoon
Administrator
Administrator


Joined: 05 Jul 2003
Posts: 54455
Location: 56N 3W

PostPosted: Thu Jan 21, 2021 12:45 am    Post subject: Reply with quote

Tony0945,

The partition numbers don't matter. The sizes matter a lot. If the data is in the correct sized partitions, you can leave it in a mess.

If the table ordering really matters, you can change it without losing any data.
A partition table is a set of pointers to data. Removing the pointers does not remove the data, it just makes it difficult to get at.

Make a note of the numbers ...
Code:
Device     Boot      Start        End    Sectors
/dev/sdc3       1644169216 1711278079   67108864
/dev/sdc4       1711278080 1953523711  242245632

Delete the partition table entries with to tool of your choice.
Make new partition table entries in the right order using *identical* numbers for the Start, End, and Sectors.
The filesystems will still be there.

Kernel SCSI device naming is not deterministic, so don't use it.
In the boot loader, use PARTUUID and in fstab use PARTUUID or filesystem UUID.

With an MSDOS partition table, the PARTUUIDs are faked and are only fixed for the four primary partitions.

If the data content is swapped, its a bit harder.
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.
Back to top
View user's profile Send private message
Tony0945
Watchman
Watchman


Joined: 25 Jul 2006
Posts: 5127
Location: Illinois, USA

PostPosted: Thu Jan 21, 2021 2:23 am    Post subject: Reply with quote

All partitions are ext4 so using labels from "e2label" in the /etc/fstab of each bootable partition.
So the grub entries should be PARTUUID if I understand you correctly. Or will labels do?

The last three partitions have copied. rsync just hung, so I mounted all the new partitions as /mnt/partN N=2,3,4 and used "cp -aur"

Now going to sign off, boot into the rescue partition and copy the main gentoo install into partition 1.
Then I'll have to:
1. edit all the /etc/fstab's using labels as noted
2. remove the SSD and carry it to a different machine. (It's actually sticking out with only the data and power cables attached.
3. Stick it in another machne, disconnect other dives to ensure it's /dev/sda. Candidate machines have bootable cdrom drives making it easier
4. boot sysrescuecd, chroot into partition1
5. run grub-install
6. for each of the other three partitions, chroot into them and install grublegacy or grub2 as appropriate
7. take the SSD out and transfer it back to the original machine.
The physical parts may be tough. It took me 15 minutes or so to get SSD mounted with the power cable correctly. Eyes are real bad today. Would up using a SATA power splitter which worked well. The existing SATA power cable has three connecters but they are short and hard to work with. Stupid pride prevented me from asking my wife to connect them.

Until tomorrow!
Back to top
View user's profile Send private message
Tony0945
Watchman
Watchman


Joined: 25 Jul 2006
Posts: 5127
Location: Illinois, USA

PostPosted: Thu Jan 21, 2021 5:29 pm    Post subject: Reply with quote

Wasted the morning stubbornly trying to get the USB usbstick to boot. Also tried running grub-install on the SSD in it's /dev/sdc position.
Then disconnected the other SATA connecters and rebooted. Endless scrolling of "GRUB Stage 1.5" message.
Along the way, I discovered that the mobo does support EFI. They just don't call it that. Tempted to scrap everything, reformat as gpt, and install r3efind. But I need a UEFI boot medium to do that. So, it's setup grub on another machine, or obtain another CD-ROM drive, possibly by disassembling a working machine. I used to have a portable CD drive from my sister's laptop. it vibrated too much to write reliably but might boot.

J
Back to top
View user's profile Send private message
Tony0945
Watchman
Watchman


Joined: 25 Jul 2006
Posts: 5127
Location: Illinois, USA

PostPosted: Fri Jan 22, 2021 11:47 pm    Post subject: Reply with quote

After the day long odyssey to restore my second system (see https://forums.gentoo.org/viewtopic-p-8563972.html#8563972). I've ordered a new DVD drive for this system.
Back to top
View user's profile Send private message
szatox
Advocate
Advocate


Joined: 27 Aug 2013
Posts: 3203

PostPosted: Sat Jan 23, 2021 11:47 am    Post subject: Reply with quote

Quote:
Tempted to scrap everything, reformat as gpt, and install r3efind. But I need a UEFI boot medium to do that.
Do you?
I know, I know, everybody says so. But why would that be the case?
Bootloader puts kernel in RAM, hands over, and leaves. What's the difference between booting in efi and legacy mode, once the main system is up and running?

I'd definitely try to put refind on that usb stick.
Because you can use another machine to do that: one that does boot.
Back to top
View user's profile Send private message
NeddySeagoon
Administrator
Administrator


Joined: 05 Jul 2003
Posts: 54455
Location: 56N 3W

PostPosted: Sat Jan 23, 2021 12:42 pm    Post subject: Reply with quote

Tony0945,

You can do an in place MSDOS to GPT migration without losing anything. There might even be a tool for it, I didn't search.
First some background, so you understand what you are getting into.

The MSDOS partition table has the primary partitions in the last 66 bytes of LBA 0.
If you have and extended partition, that reserves space and points to the first entry in a linked list that defines the logical partitions.
Each logical partition contains a pointer to the next logical partition, if there is one. Hence when you delete partition 5, all the rest move down one.

GPT has a 'protective, MSDOS partition table in LBA 0 then the first copy of the GPT follows. It occupies LBA 1 to 2047. Hence partitioning tools default to starting the first partition at LBA 2048.
So you have space for the first copy of the GPT before the first partition.
GPT keeps a second copy at the end of the block device too. That's unfortunate as with MSDOS in use you will likely have a filesystem there.

The method then, is to shrink the filesystem at the physical end of the device to make room for the backup copy of the GPT.
Copy the sector numbers from the existing MSDOS partition table.
Write a new GPT disk label. If you missed the filesystem shrink, this step will trash that filesystem.
Put the old partition boundaries back. Including the extended partition if there was one.
WARNING:the physically last partition on the device is now about 2048 blocks smaller as the space has been taken up by the GPT copy, so make the partition as big as you can.
Everything should just work but you did make a backup before you started, didn't you?

Grow the final filesystem to fit the space in its new home.
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.
Back to top
View user's profile Send private message
Tony0945
Watchman
Watchman


Joined: 25 Jul 2006
Posts: 5127
Location: Illinois, USA

PostPosted: Mon Jan 25, 2021 2:31 am    Post subject: Reply with quote

Yes, NeddySeagoon, I could do that. There is plenty of room to shrink the last partition. But copying from the existing HDD was easy, just took time. I'd probably be better off re-partitioning it totally and putting a swap partition back. while editing the fstab's I see that the 32bit partition and the RESCUE partition use the existing swap device. Not sure about the Ubuntu partition. it might be best to just re0install Ubuntu anyway. I have no files there. My main Gentoo partition used both the external swap partition and a 12G swapfile, so deleting the swap partition was no problem. I could do that for the rescue partition also. Not sure about the 32bit partition. It should have swap as it's a build box for the k6 machine. The PC has 8G RAM with all slots full. But the 32bit install can only see half of it. 4G is not enough for some packages.

My new CD drive came yesterday. Now to find time to install it.
Back to top
View user's profile Send private message
timeBandit
Bodhisattva
Bodhisattva


Joined: 31 Dec 2004
Posts: 2719
Location: here, there or in transit

PostPosted: Tue Jan 26, 2021 7:31 pm    Post subject: Reply with quote

Out of curiosity: Why allocate so much space to the rescue partition? For something that's truly for emergency use only, 145GB seems awfully chunky. I'd bet there's another 100GB there for the taking, to allocate to your other partitions, without (I would think) seriously compromising rescue functionality.

For any binary distro I wouldn't have thought twice but you did say, it's a Gentoo rescue partition. So why not make it lean and mean?

To illustrate, my small home server--with Apache, Postfix+Dovecot, a bare-bones DLNA server, a minimal Xfce desktop, and more besides--has an installed footprint of ~13GB, about 20% overall utilization of the system partitions. (Data reside on a separate, much larger HDD). Granted, this is a 32-bit system, and perhaps unattainably compact in a 64-bit world. I'd still bet 50% utilization of a 40-50GB partition is attainable with a solid collection of fix-my-system tools and a trim-but-useful desktop environment...and there's your 100GB.

Of course, "not worth the trouble to prune it" is a perfectly valid choice, too. :)
_________________
Plants are pithy, brooks tend to babble--I'm content to lie between them.
Super-short f.g.o checklist: Search first, strip comments, mark solved, help others.
Back to top
View user's profile Send private message
Tony0945
Watchman
Watchman


Joined: 25 Jul 2006
Posts: 5127
Location: Illinois, USA

PostPosted: Tue Jan 26, 2021 8:13 pm    Post subject: Reply with quote

timeBandit wrote:
Out of curiosity: Why allocate so much space to the rescue partition? For something that's truly for emergency use only, 145GB seems awfully chunky. I'd bet there's another 100GB there for the taking, to allocate to your other partitions, without (I would think) seriously compromising rescue functionality.

I'm sure you are right! The real reason is that's the space that was left and I can only have four partitions without screwing with an extended partition. I'm considering blowing away the Ubuntu partition. I only have it to build a certain software that the developer swears builds on Ubuntu but throws errors on Gentoo. I think a better option is to debug his build script.
Back to top
View user's profile Send private message
Tony0945
Watchman
Watchman


Joined: 25 Jul 2006
Posts: 5127
Location: Illinois, USA

PostPosted: Wed Jan 27, 2021 8:34 pm    Post subject: Reply with quote

All right! halfway home. I procured a new CDROM drive and still couldn't boot it. Tried setting BIOS from EFI MODE -Auto to EFI MODE - Bios. Not go.
Eventually found that setting SATA ports 4/5 to IDE instead of SATA (even though drive IS SATA) worked to get me into sysresecue CD.

From there, I chrooted into /devsda1 and ran "grub-install /dev/sda" (grub legacy is installed), then rebooted.
Success! (I thought). My familiar grub menu appeared. but when I selected the first entry, it failed to load with, I think, "can't find root device".
Right now I'm in the chroot, the only drives are the SSD at SATA port 0 and the LITE-ON CD drive at SATA port 4
Code:
~ $ ls /dev/sd*
/dev/sda  /dev/sda1  /dev/sda2  /dev/sda3  /dev/sda4


Code:
~ $ cat /boot/grub/grub.conf
default 0
timeout 10
splashimage=(hd0,0)/boot/grub/splash.xpm.gz

title=Gentoo Next Latest (5.4.72-gentoo)
root (hd0,0)
kernel /vmlinuz-5.4.72-gentoo vga=0x346 rootfstype=ext4 root=/dev/sda1 net.ifnames=0 mitigations=off acpi_enforce_resources=lax

title=Gentoo (4.17.19-gentoo)
root (hd0,0)
kernel /vmlinuz-4.17.19-gentoo vga=0x346 rootfstype=ext4 root=/dev/sda1 net.ifnames=0 mitigations=off

title=32bit
root (hd0,1)
chainloader +1

title=Ubuntu
root (hd0,2)
chainloader +1

title=RESCUE GENTOO
root (hd0,3)
chainloader +1
Neither kernel boots.

Code:
 $ uname -a
Linux Casti 4.14.80-std532-amd64 #2 SMP Sun Nov 11 18:59:44 UTC 2018 x86_64 AMD Phenom(tm) II X6 1090T Processor AuthenticAMD GNU/Linux
WTF? I guess this is sysrescuecd's kernel. Doh! Of course it is.
And here is what the kernel that grub legacy tried to boot has"
Code:
 $ grep CMDLINE /boot/config-5.4.72-gentoo
CONFIG_CMDLINE_BOOL=y
CONFIG_CMDLINE="net.ifnames=0 mitigations=off"
# CONFIG_CMDLINE_OVERRIDE is not set
# CONFIG_BLK_CMDLINE_PARSER is not set
# CONFIG_CMDLINE_PARTITION is not set
CONFIG_FB_CMDLINE=y

Will attempt to boot by UUID next.
EDIT:
Hadn't expected the system clock to be so far off:
Code:
sudo /usr/sbin/ntpdate pool.ntp.org
27 Jan 14:37:31 ntpdate[3804]: step time server 216.218.254.202 offset +21600.605901 sec

EDIT2:
Code:
~ $ e2label /dev/sda
e2label: Bad magic number in super-block while trying to open /dev/sda
/dev/sda contains `DOS/MBR boot sector; GRand Unified Bootloader, stage1 version 0x3, stage2 address 0x2000, stage2 segment 0x200' data
Doesn't look good. I seem to have lost the labels.
EDIT3:
Code:
~ $ blkid
/dev/loop0: TYPE="squashfs"
/dev/sda1: LABEL="MX500ROOT" UUID="72fffe1a-89f5-4b49-9de7-2f37b12e923f" BLOCK_SIZE="4096" TYPE="ext4" PARTUUID="efbea596-01"
/dev/sda2: LABEL="32BIT" UUID="e6f583cf-61aa-474b-88ab-181fda2bf304" BLOCK_SIZE="4096" TYPE="ext4" PTTYPE="dos" PARTUUID="efbea596-02"
/dev/sda3: LABEL="UBUNTU" UUID="0ace029a-b5b0-460c-a39f-bc419c8035bc" BLOCK_SIZE="4096" TYPE="ext4" PARTUUID="efbea596-03"
/dev/sda4: LABEL="RESCUE" UUID="c6cd5b2d-7ece-4bc6-abf8-0f97bc8c3b8c" BLOCK_SIZE="4096" TYPE="ext4" PARTUUID="efbea596-04"
/dev/sr0: BLOCK_SIZE="2048" UUID="2018-11-14-16-35-14-00" LABEL="sysrcd-5.3.2" TYPE="iso9660"

Labels are there. Will try a boot stanza using PARTUUID.
Back to top
View user's profile Send private message
Tony0945
Watchman
Watchman


Joined: 25 Jul 2006
Posts: 5127
Location: Illinois, USA

PostPosted: Wed Jan 27, 2021 10:38 pm    Post subject: Reply with quote

PARTUUID didn't work either. this time I wrote the error message down.

"ERROR 15: File not found." I think it's a grub message.

I chrooted into the 32bit partition using linux32. I re-installed grub-legacy with "grub-install /dev/sda2". It failed to chain load with the same ERROR 15.
Back to top
View user's profile Send private message
NeddySeagoon
Administrator
Administrator


Joined: 05 Jul 2003
Posts: 54455
Location: 56N 3W

PostPosted: Wed Jan 27, 2021 11:35 pm    Post subject: Reply with quote

Tony0945,

info grub:
15 : File not found
     This error is returned if the specified file name cannot be found,
     but everything else (like the disk/partition info) is OK.

Thats from grub stage2.

Code:
root (hd0,0)
kernel /vmlinuz-5.4.72-gentoo

That says to load the file vmlinuz-5.4.72-gentoo from the top level of (hd0,0), which is sda1.
Are you sure you don't need
Code:
kernel /boot//vmlinuz-5.4.72-gentoo

_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.
Back to top
View user's profile Send private message
Tony0945
Watchman
Watchman


Joined: 25 Jul 2006
Posts: 5127
Location: Illinois, USA

PostPosted: Thu Jan 28, 2021 1:50 am    Post subject: Reply with quote

NeddySeagoon wrote:

Are you sure you don't need
Code:
kernel /boot//vmlinuz-5.4.72-gentoo

Yes! I'm a dunce. The original disk layout had a separate /boot. When grub reads it, all the kernels are at the top of the partition. Hence, the grub.conf called them /vmlinuz... They didn't become /boot/vmlinuz-.... until the newly loaded kernel mounts the root partition.
Now that there is only one partition, the grub line should say /boot/vmlinuz-... as you say. Once the kernel is loaded and jumped into, the kernel is in memory and knows nothing of what grub knew or did. So the internal command line should be OK. That command line doesn't say where root is. I'm guessing that it's default is to assume hd(0,0) from the error messages I've seen on the forum. I'm hoping that message isn't literal and the kernel assumes the partition that it's loaded from. Either that or it accepts the command line grub passed in.

The 32bit partition didn't have a separate /boot before either and grub was installed on the partition. I re-installed it from a chroot but it didn't boot either. Maybe the same problem? I don't use it often and as long as I have one booting Linux partition, I can always chroot. Those are the two grub-legacy booters. The other two are grub2. I'll have to look up that partition boot command for grub2 again. I need a new filing system!
Alas tomorrow is shopping day. After that I'll open up the box and make the grub.conf change.

I'm really tempted to try EFI and refind on this machine, but I'm afraid of the flaky BIOS. I have to tell it the SATA DVD is IDE? That's dumb!
I also ordered a few more Molex to SATA power adapters. Always good when working on old machines. To tell the truth, I prefer Molex. it's much easier to attach.
Back to top
View user's profile Send private message
Tony0945
Watchman
Watchman


Joined: 25 Jul 2006
Posts: 5127
Location: Illinois, USA

PostPosted: Thu Jan 28, 2021 2:04 am    Post subject: Reply with quote

BTW, I did run "hdparm -tT /dev/sda" on the SSD when in the chroot. 532Mbps. Quite sweet! Fantastic to a guy that started with eight inch floppys on an Intel develpment system.
Back to top
View user's profile Send private message
Tony0945
Watchman
Watchman


Joined: 25 Jul 2006
Posts: 5127
Location: Illinois, USA

PostPosted: Thu Jan 28, 2021 6:35 pm    Post subject: Reply with quote

Up and running and booting on it's own!
Major problem with booting this morning. Would not boot from CD and could not open cd drive.
I unhooked every other drive and sysrescuecd 5.3.2 booted promptly. Threw away the old SATA cables.
BIOS still flaky about entering. Now chokes on DEL, halfway through listing devices. Still responds to F12 to select boot device. So I've left it at "Boot from CDROM first" If I don't have a bootable medium there, it switches to the first HDD. I'm thinking this BIOS need to be re-flashed but I'm not going to go there now.

Both grub legacy partitions are booting. The grub2 rescue partition was rescued by a procedure that I had to search for and put as a note in /boot/grub
Code:
 # cat HOWTOPARTITIONINSTALL
#example install into partition 4 of drive sda
# grub-install /dev/sda4 --force
# next command will update the menu
#  grub-mkconfig -o /boot/grub/grub.cfg


As promised here is the read speed on this Crucial MX500 series 1TB drive.
Code:
# hdparm -tT /dev/sda

/dev/sda:
 Timing cached reads:   6836 MB in  2.00 seconds = 3418.79 MB/sec
 Timing buffered disk reads: 1584 MB in  3.00 seconds = 527.49 MB/sec
That's 3.67 X as fast as the 10,000 RPM Velociraptor. I can hardly wait to see what an NVME drive can do! Not to mention no more hunting to shift tracks.
I see there are 4TB SSD'sfor sale. Stratospheric prices. I have a new 4TB WD Gold. That was pricey enough.

Nothing left to do now but put the two drives in the drive cage (now sprawling across the floor net to the case) and set the last two DVD drives aside for further investigation on another machine. They really shouldn't be bad. Crossing my fingers because my old k6 boots from an IDE CDROM and an incredibly old WD Black IDE drive.

Didn't try to mount the Ubuntu partotion yet. Maybe I should just update it to a newer UBUNTU. Maybe I should just forget it and run an Ubuntu VM on the ryzen box.

Is there anything I've neglected to say?
Back to top
View user's profile Send private message
NeddySeagoon
Administrator
Administrator


Joined: 05 Jul 2003
Posts: 54455
Location: 56N 3W

PostPosted: Thu Jan 28, 2021 7:17 pm    Post subject: Reply with quote

Tony0945,

root (hd0,0) is grub speak for where to find the files that grub will load.
root= points tho the kernels root partition.

The two uses of the word root mean two different things.

It appears that's the hard bit done now that it boots on its own.

You can use the kernel command line
Code:
kernel /booot/somekernel ...
regardless of the existence of a boot partition or not.
Put a symlink
Code:
boot -> .
at the top level of /boot.
That soaks up the extra /boot if its not needed and gets ignored if its not required.
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.
Back to top
View user's profile Send private message
saellaven
l33t
l33t


Joined: 23 Jul 2006
Posts: 648

PostPosted: Thu Jan 28, 2021 7:29 pm    Post subject: Reply with quote

Tony0945 wrote:

As promised here is the read speed on this Crucial MX500 series 1TB drive.
Code:
# hdparm -tT /dev/sda

/dev/sda:
 Timing cached reads:   6836 MB in  2.00 seconds = 3418.79 MB/sec
 Timing buffered disk reads: 1584 MB in  3.00 seconds = 527.49 MB/sec
That's 3.67 X as fast as the 10,000 RPM Velociraptor. I can hardly wait to see what an NVME drive can do! Not to mention no more hunting to shift tracks.
I see there are 4TB SSD'sfor sale. Stratospheric prices. I have a new 4TB WD Gold. That was pricey enough.


Code:

# hdparm -tT /dev/nvme0n1

/dev/nvme0n1:
 Timing cached reads:   28132 MB in  2.00 seconds = 14084.53 MB/sec
 Timing buffered disk reads: 9112 MB in  3.00 seconds = 3036.69 MB/sec

# smartctl -a /dev/nvme0n1
smartctl 7.1 2019-12-30 r5022 [x86_64-linux-5.10.9] (local build)
Copyright (C) 2002-19, Bruce Allen, Christian Franke, www.smartmontools.org

=== START OF INFORMATION SECTION ===
Model Number:                       Samsung SSD 970 EVO Plus 500GB


I paid $80 for this a month ago.
_________________
Ryzen 3700X, Asus Prime X570-Pro, 64 GB DDR4 3200, GeForce GTX 1660 Super
openrc-0.17, ~vanilla-sources, ~nvidia-drivers, ~gcc
Back to top
View user's profile Send private message
saellaven
l33t
l33t


Joined: 23 Jul 2006
Posts: 648

PostPosted: Thu Jan 28, 2021 7:53 pm    Post subject: Reply with quote

oh, and for a comparison to SSD on the same machine

Code:

# hdparm -tT /dev/sdb

/dev/sdb:
 Timing cached reads:   26606 MB in  2.00 seconds = 13318.96 MB/sec
 Timing buffered disk reads: 1630 MB in  3.00 seconds = 542.82 MB/sec

# smartctl -a /dev/sdb
smartctl 7.1 2019-12-30 r5022 [x86_64-linux-5.10.9] (local build)
Copyright (C) 2002-19, Bruce Allen, Christian Franke, www.smartmontools.org

=== START OF INFORMATION SECTION ===
Model Family:     Samsung based SSDs
Device Model:     Samsung SSD 860 EVO 1TB

_________________
Ryzen 3700X, Asus Prime X570-Pro, 64 GB DDR4 3200, GeForce GTX 1660 Super
openrc-0.17, ~vanilla-sources, ~nvidia-drivers, ~gcc
Back to top
View user's profile Send private message
Tony0945
Watchman
Watchman


Joined: 25 Jul 2006
Posts: 5127
Location: Illinois, USA

PostPosted: Thu Jan 28, 2021 8:06 pm    Post subject: Reply with quote

saellaven,
Awesome speed. Awesome price!

542, 527. Pretty close. Velociraptor rotated nearly twice as fast as a standard hard disk.
Back to top
View user's profile Send private message
Display posts from previous:   
Reply to topic    Gentoo Forums Forum Index Other Things Gentoo All times are GMT
Goto page Previous  1, 2, 3  Next
Page 2 of 3

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum