-
Notifications
You must be signed in to change notification settings - Fork 76
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Best strategy for preparing multiple IIAB servers using imaging #3247
Comments
Perhaps you could share your understanding of what is needed and the tools that you are going to use to support your endeavor. My guidance is recorded in #3106, 'b' is not a problem if you make both data and os 'images' at the same time from the source machine. |
not a problem. this is the git repo and will be present as part of the install. it is also not used after the install. I think we have come down to two strategies:
@jvonau says 2) is not viable because df will return the wrong or no size. If that is true then 1) should be used. The only downside is that LIBRARY-EXT will be usable in Admin Console Manage Content as an internal device, but not as an external device mountable on another system. |
Ideally we will have 2 separate images for /library and OS/IIAB |
Quick Recap from our weekly community call (http://minutes.iiab.io) ~8 hours ago. @tim-moody's post from June 4th appears to be the most immediately useful here: (for this ticket especially) To reconfirm that the various
@jvonau had suggestions regarding future use of a loopback file (loopback mount?) towards possibly streamlining [*] Not quite a Learning Bouquet or "preset", but a very similar idea! Finally, we did not have time to complete the discussion, as there are several quite subtle + nuanced issues to review regarding what belongs "externally" on /library — versus what belongs "internally" on IIAB root filesystems — to be discussed on a future call. |
@tim Moody ***@***.***>
@jvonau <https://github.com/jvonau>
Guys I was testing this on my fresh Mint 20.3 install on a laptop with a
750GB HDD as discussed and encountered problems perhaps due to my
misunderstanding of the sequence required...
I used Gparted to re-partiton the disk as per screenshot attached and
labelled /dev/sda6 to take the library, edited /etc/fstab as per finally
agreed version then
***@***.***:~$ sudo mount -a
[sudo] password for user1:
mount: /library-ext: mount point does not exist.
mount: /library: mount point does not exist.
***@***.***:~$ lsblk
NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT
sda 8:0 0 698.7G 0 disk
├─sda1 8:1 0 513M 0 part /boot/efi
├─sda2 8:2 0 512M 0 part /media/user1/055D-46BB
├─sda3 8:3 0 1K 0 part
├─sda5 8:5 0 15.4G 0 part /
├─sda6 8:6 0 488.9G 0 part
└─sda7 8:7 0 193.4G 0 part
sdf 8:80 1 3.8G 0 disk
└─sdf1 8:81 1 3.8G 0 part
sr0 11:0 1 1024M 0 rom
***@***.***:~$ blkid
/dev/sda1: UUID="7616-1C9B" TYPE="vfat" PARTUUID="bc55cb24-01"
/dev/sda2: UUID="055D-46BB" TYPE="vfat" PARTUUID="bc55cb24-02"
/dev/sda5: UUID="df13e79b-06b1-468f-9660-27157428e78c" TYPE="ext4"
PARTUUID="bc55cb24-05"
/dev/sda6: LABEL="IIAB-LIBRARY-EXT"
UUID="581d987e-7435-4709-a25f-e9d804f170e1" TYPE="ext4"
PARTUUID="bc55cb24-06"
/dev/sda7: LABEL="Backup" UUID="ab3e4b96-cffc-4558-a748-6f72f0210ea0"
TYPE="ext4" PARTUUID="bc55cb24-07"
/dev/sdf1: UUID="BCC8-B24E" TYPE="vfat"
(last line is just a USB with the procedure docs on it)
So where am I going wrong here?
…On Thu, Jun 9, 2022 at 11:48 PM A Holt ***@***.***> wrote:
Quick Recap from our weekly community call (http://minutes.iiab.io) than
began 8h ago.
@tim-moody <https://github.com/tim-moody>'s post from June 4th appears to
be the most immediately useful here: (for this ticket especially)
- #3106 (comment)
<#3106 (comment)>
To reconfirm that the various df issues are all under control,
@shanti-bhardwa <https://github.com/shanti-bhardwa> and @andrewclark5
<https://github.com/andrewclark5> will try this out to verify that it
works well on Mint 20.3+ on the hardware most relevant for Turing Trust —
verifying that it allows for:
1. Easily mountable (easily interchangeable) external IIAB content
disks.
2. With Admin Console's (http://box.lan/admin) > Install Content > Manage
Content
<https://github.com/iiab/iiab-admin-console/blob/master/roles/console/files/help/InstContent.rst#manage-content>
continuing to work.
@jvonau <https://github.com/jvonau> had suggestions regarding future use
of a loopback file (loopback mount?) towards possibly streamlining
/library content drives/directories/etc in future (much like a partition,
but not quite the same). For example in cases where 100 or 128 GB of
/library content has already been decided upon, and the inevitable
last-minute content changes calm down, allowing this "content collection"
[*] to be circulated among many similar IIAB's?
[*] Not quite Learning Bouquet or "preset", but a very similar idea!
Finally, we did not have time to complete the discussion, as there are
several quite subtle + nuanced issues to review regarding what belongs
"externally" on /library — versus what belongs "internally" on IIAB root
filesystems. To be discussed on a future call.
—
Reply to this email directly, view it on GitHub
<#3247 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AZLTUR6T3XVRSLEFQIX2ANDVOJYD7ANCNFSM5YJZ7ZFA>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
instructions have gotten spread over so many locations that things easily go missing.
as the error messages indicate you have skipped this step
now mount -a should work I take it that using an internal partition is just an experiment and not the eventually intended configuration. |
/dev/sda6 also needs a library directory on it before mount -a will work |
OK. I wondered about that. But in order to create the directories on that
partition I first have to mount it, right?
That was what I was attempting to do, with the mount - a command.
I take it that using an internal partition is just an experiment and not
the eventually intended configuration.
Not exactly. This is what I was saying yesterday - if that WOULD work for
big disks then we would have a standardization of procedure across all
server builds, big or small..
…On Fri, Jun 10, 2022 at 1:18 PM Tim Moody ***@***.***> wrote:
instructions have gotten spread over so many locations that things easily
go missing.
mount: /library-ext: mount point does not exist.
mount: /library: mount point does not exist.
as the error messages indicate you have skipped this step
mkdir /library
mkdir /library-ext
now mount -a should work
I take it that using an internal partition is just an experiment and not
the eventually intended configuration.
—
Reply to this email directly, view it on GitHub
<#3247 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AZLTURYY5LLKLWY5B6NNHOTVOMXABANCNFSM5YJZ7ZFA>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
if the data (library) partition is internal I would think it would not ever be mounted externally on another system if that is the case then the bind mount is not really needed and you can mount /dev/sda6 directly on /library in fstab and comment out the other bind mount line all of my nuc machines are configured like this |
I should add that the bind mount will still work even for an internal partition if standardization is the top priority. |
Looks as if we are partly there, after a reboot. (So sudo mount -a is not
the same as rebooting, I guess?)
/dev/sda6 is now mounted, but looks as if I need to change it to /library
and not /library-ext, which is what is appearing in the File System?
So I will now try
LABEL=IIAB-LIBRARY-EXT /library ext4 defaults,nofail 0 0
# was /library-ext/library /library none defaults,bind,nofail 0 0
…On Fri, Jun 10, 2022 at 1:42 PM Tim Moody ***@***.***> wrote:
I should add that the bind mount will still work even for an internal
partition if standardization is the top priority.
—
Reply to this email directly, view it on GitHub
<#3247 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AZLTUR3OOIA4W7LR4XPDR4DVOMZ2XANCNFSM5YJZ7ZFA>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
So now it looks like you have unexpectedly switched strategies to mounting the internal partition directly on /library. Let's get the main strategy working first. with /dev/sda6 mounted on /library-ext with fstab as LABEL=IIAB-LIBRARY-EXT /library-ext ext4 defaults,nofail 0 0 mount -a |
OK, did all that.
***@***.***:~$ sudo nano /etc/fstab
[sudo] password for user1:
***@***.***:~$ sudo mount -a
***@***.***:~$ blkid
/dev/sda1: UUID="7616-1C9B" TYPE="vfat" PARTUUID="bc55cb24-01"
/dev/sda2: UUID="055D-46BB" TYPE="vfat" PARTUUID="bc55cb24-02"
/dev/sda5: UUID="df13e79b-06b1-468f-9660-27157428e78c" TYPE="ext4"
PARTUUID="bc55cb24-05"
/dev/sda6: LABEL="IIAB-LIBRARY-EXT"
UUID="581d987e-7435-4709-a25f-e9d804f170e1" TYPE="ext4"
PARTUUID="bc55cb24-06"
/dev/sda7: LABEL="Backup" UUID="ab3e4b96-cffc-4558-a748-6f72f0210ea0"
TYPE="ext4" PARTUUID="bc55cb24-07"
***@***.***:~$
***@***.***:~$ lsblk
NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT
sda 8:0 0 698.7G 0 disk
├─sda1 8:1 0 513M 0 part /boot/efi
├─sda2 8:2 0 512M 0 part
├─sda3 8:3 0 1K 0 part
├─sda5 8:5 0 15.4G 0 part /
├─sda6 8:6 0 488.9G 0 part /library
└─sda7 8:7 0 193.4G 0 part
sr0 11:0 1 1024M 0 rom
***@***.***:~$
Is that what you want?
…On Fri, Jun 10, 2022 at 2:33 PM Tim Moody ***@***.***> wrote:
/dev/sda6 is now mounted, but looks as if I need to change it to /library
and not /library-ext, which is what is appearing in the File System?
So now it looks like you have unexpectedly switched strategies to mounting
the internal partition directly on /library.
Let's get the main strategy working first.
with /dev/sda6 mounted on /library-ext
mkdir /library-ext/library
with fstab as
LABEL=IIAB-LIBRARY-EXT /library-ext ext4 defaults,nofail 0 0
/library-ext/library /library none defaults,bind,nofail 0 0
mount -a
—
Reply to this email directly, view it on GitHub
<#3247 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AZLTUR6ZNES2TFKBJFO5FJLVOM73NANCNFSM5YJZ7ZFA>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
...but I have now ended up with a directory called library-ext containing
sub-directories for library and library-ext!
…On Fri, Jun 10, 2022 at 2:33 PM Tim Moody ***@***.***> wrote:
/dev/sda6 is now mounted, but looks as if I need to change it to /library
and not /library-ext, which is what is appearing in the File System?
So now it looks like you have unexpectedly switched strategies to mounting
the internal partition directly on /library.
Let's get the main strategy working first.
with /dev/sda6 mounted on /library-ext
mkdir /library-ext/library
with fstab as
LABEL=IIAB-LIBRARY-EXT /library-ext ext4 defaults,nofail 0 0
/library-ext/library /library none defaults,bind,nofail 0 0
mount -a
—
Reply to this email directly, view it on GitHub
<#3247 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AZLTUR6ZNES2TFKBJFO5FJLVOM73NANCNFSM5YJZ7ZFA>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
Having to create /library-ext/library forces editing /etc/fsab and using mount -a twice adding the two /etc/fstab entries individually. I've baked this functionality into #3051 |
Hi Jerry That seems like a good explanation! I am talking to Tim on email just now as he is out so I see you have cc your text to him by email. |
Should I delete all the superfluous folders and start again with fstab?
And what are we trying to achieve with the library folder structure?
…On Fri, Jun 10, 2022 at 3:38 PM Jerry Vonau ***@***.***> wrote:
Having to create /library-ext/library forces editing /etc/fsab and using
mount -a twice adding to two /etc/fstab entries individually.
First edit to add the mounting of /library-ext line.
sudo mount -a
sudo mkdir /library-ext/library
Second edit to fstab adding the bindmount line.
sudo mount -a
I've baked this functionality into #3051
<#3051>
—
Reply to this email directly, view it on GitHub
<#3247 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AZLTUR224B2T4Z55QYZYYEDVONHPNANCNFSM5YJZ7ZFA>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
On the OS filesystem there should be two folders /library /library-ext
Think Tim's goal is to avoid rewriting current code in admin-console to be able to extend the existing import/export functionality. |
@tim-moody @jvonau @holta |
Done some testing on HP t620 with 16GB flash drive. Results look promising. Here is a summary.
|
I think this going work really well. You can see the amount of storage on the uSB drives from Manage Content option. |
Trying to get System Storage on Admin Console gives me this error |
Apologies, the machine is on vpn 10.8.0.62. My user id is sb01 with password shanti. Also iiab-admin password is shanti. |
Next steps for me to test is to create more External drives. So I will clone the IIAB-LIBRARY-EXT. |
If you use gparted to look at the drives mounted then IIAB-LIBRARY-EXT which is /dev/sdb1 is mounted in three places /library, /library-ext and /media/usb0. You also get the used and unused storage size. So there must be other ways to get storage information without using df command. |
even after reboot? (leaving it attached)
yes it's a bug - not compatible with df change for linux mint. |
Shanti
For reference
Shanti version - on USB
LABEL=IIAB-LIBRARY-EXT /library-ext ext4 defaults,nofail 0 0
/library-ext/library /library none defaults,bind,nofail, 0 0
Current HP g6 version - on HDD
# Added for IIAB Library - here using HDD partition /dev/sda6
# Not currently working - puts /library on sda5 File System
LABEL=IIAB-LIBRARY-EXT /library-ext ext4 defaults,nofail 0 0
/library-ext/library /library none defaults,bind,nofail 0 0
As far as I can see they are the same syntax but being treated differently
due to me using an internal partition in an effort to standardise. But
***@***.***:~$ df
df: /run/user/1000/doc: Operation not permitted
Filesystem 1K-blocks Used Available Use% Mounted on
udev 1696672 0 1696672 0% /dev
tmpfs 348688 3120 345568 1% /run
/dev/sda5 14793648 11812176 2160328 85% /
tmpfs 1743432 0 1743432 0% /dev/shm
tmpfs 5120 4 5116 1% /run/lock
tmpfs 1743432 0 1743432 0% /sys/fs/cgroup
/dev/sda6 503578816 1488376 476440420 1% /library-ext
/dev/sda1 524272 4 524268 1% /boot/efi
tmpfs 348684 24 348660 1% /run/user/1000
Andrew
…On Sat, Jun 11, 2022 at 1:34 PM Tim Moody ***@***.***> wrote:
df -h shows the usb drive mounted on /media/usb0
even after reboot?
GET-STORAGE-INFO FAILED
yes it's a bug - not compatible with df change for linux mint.
—
Reply to this email directly, view it on GitHub
<#3247 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AZLTURYEBI3WKQSYV4QYYE3VOSBUHANCNFSM5YJZ7ZFA>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
I don't know if this throws any further light on what is happening on my HP
g6
***@***.***:/$ cd library
***@***.***:/library$ ls
awstats calibre-web downloads ka-lite kolibri public Test.txt
wordpress working www zims
***@***.***:/library$ cd /
***@***.***:/$ ls -l /lib*
lrwxrwxrwx 1 root root 7 Jun 9 13:29 /lib -> usr/lib
lrwxrwxrwx 1 root root 9 Jun 9 13:29 /lib32 -> usr/lib32
lrwxrwxrwx 1 root root 9 Jun 9 13:29 /lib64 -> usr/lib64
lrwxrwxrwx 1 root root 10 Jun 9 13:29 /libx32 -> usr/libx32
/library:
total 40
drwxr-x--- 2 www-data www-data 4096 Jun 11 13:40 awstats
drwxr-xr-x 3 root www-data 4096 Jun 10 19:15 calibre-web
drwxr-xr-x 5 root root 4096 Jun 10 18:56 downloads
drwxr-xr-x 9 root root 4096 Jun 11 11:23 ka-lite
drwxr-xr-x 6 kolibri www-data 4096 Jun 11 13:47 kolibri
drwxrwxrwx 2 smbuser smbuser 4096 Jun 10 19:03 public
-rw-r--r-- 1 root root 0 Jun 10 17:40 Test.txt
drwxrwxr-x 5 root www-data 4096 Jun 10 19:06 wordpress
drwxr-xr-x 6 root root 4096 Jun 10 19:18 working
drwxr-xr-x 4 root root 4096 Jun 10 19:11 www
drwxr-xr-x 4 root root 4096 Jun 10 19:22 zims
/library-ext:
total 24
drwxr-xr-x 12 root root 4096 Jun 10 19:14 library
drwx------ 2 root root 16384 Jun 10 12:03 lost+found
-rw-r--r-- 1 root root 72 Jun 10 14:31 TestDoc.txt
***@***.***:/$
I can start the VPN support if it helps.
Andrew
…On Sat, Jun 11, 2022 at 1:34 PM Tim Moody ***@***.***> wrote:
df -h shows the usb drive mounted on /media/usb0
even after reboot?
GET-STORAGE-INFO FAILED
yes it's a bug - not compatible with df change for linux mint.
—
Reply to this email directly, view it on GitHub
<#3247 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AZLTURYEBI3WKQSYV4QYYE3VOSBUHANCNFSM5YJZ7ZFA>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
Think this point to the use of 'oem' install or the video subsystem as the primary cause of the 'crash' |
I can run the procedure on it to start as a new user if you like?
It will need then to be rebooted and the VPN reconnected.
Is there any specific indication that this is causing the issue?
Regards
Andrew
Sent from my Moto 5G
Sent from my Google Mail account
…On Mon, 27 Jun 2022, 7:12 pm Jerry Vonau, ***@***.***> wrote:
Sure. Will do both.
Should add the IIAB machine display is stable again. I hate random stuff
like this!
Would you believe that the "clean" Mint-only T620-2 is now exhibiting the
same blacking-out screen bug! Here is the output from .xsession-errors
Think this point to the use of 'oem' install or the video subsystem as the
primary cause of the 'crash'
—
Reply to this email directly, view it on GitHub
<#3247 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AZLTURYBOUZD6W5ZEYC4PYLVRHVG5ANCNFSM5YJZ7ZFA>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
I keep wondering if this is a timing problem: some cinnamon dependency not being available in time for auto login. Maybe more likely on IIAB as there are more services to start at boot. |
Instructed to proceed through the oem phase which will prompt on next boot to create new user/password, this is going to be the shipping state of the device. After new user creation, no complaints about desktop glitches yet, left to run overnight.
|
Seem ok poking around but kolibri's cpu usage seems a bit high. Is that 'out of the box' or are there channels loaded?
|
I was testing Kolibri last night and exploring one of the channels we
created for the Malawi IT curriculum. Would that have done it?
In fact I tested a lot of the content of my 25GB Small Library. Most of it
works but some modules need fixing as discussed on our call last week.
Whilst the system seems stable now that I have run the System Preparation
tool as a new user I will continue testing today and report results. 😀
…On Tue, 28 Jun 2022, 8:40 am Jerry Vonau, ***@***.***> wrote:
Seem ok poking around but kolibri's cpu usage seems a bit high. Is that
'out of the box' or are there channels loaded?
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
1424 kolibri 20 0 2079496 82412 7156 S 16.9 2.4 103:03.41 kolibri
794 root 20 0 773856 115236 77736 S 1.0 3.4 1:47.87 Xorg
11406 user1 20 0 11888 3924 3420 R 1.0 0.1 0:00.11 top
3773 user1 20 0 806860 66728 46220 S 0.7 1.9 0:17.89 cinnamon-screen
828 mysql 20 0 1710492 78504 18592 S 0.3 2.3 1:06.55 mysqld
1025 root 20 0 1202568 42968 5424 S 0.3 1.2 1:20.85 kalite
1953 user1 20 0 3773524 176176 107088 S 0.3 5.1 2:43.00 cinnamon
—
Reply to this email directly, view it on GitHub
<#3247 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AZLTUR36DUCIYXHQS5GFQH3VRKUARANCNFSM5YJZ7ZFA>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
I just rebooted and checked the running processes again. It does look as if Kolibri is a bit resouce-hungry as there are no clients running IIAB. top - 09:20:50 up 9 min, 1 user, load average: 1.18, 1.34, 0.84
|
Definitely. This is a known issue. Kolibri uses significant resources even when idle. While this doesn't break the bank, certainly there are many who find this aggravating, and it's something to be aware of (-: |
That's useful to know as we have committed quite a lot of time to
developing Kolibri for the Turing Trust educational sites in Malawi.
…On Tue, 28 Jun 2022, 9:28 am A Holt, ***@***.***> wrote:
I just rebooted and checked the running processes again. It does look as
if Kolibri is a bit resouce-hungry as there are no clients running IIAB.
Definitely. This is a known issue. Kolibri uses significant resources even
when idle.
While this doesn't break the bank, certainly there are many who find this
aggravating, and it's something to be aware of (-:
—
Reply to this email directly, view it on GitHub
<#3247 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AZLTUR3P4PAYCVTJQZEN5WTVRKZTBANCNFSM5YJZ7ZFA>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
OK I have been hammering this machine all day with reboots and downloads of RACHEL modules without a glitch. It seems @jvonau Jerry was probably right and it was the OEM build option that pushed it over the edge! We may need to be careful in future about adding any further load to the desktop system as I deduce that it might be a bit fragile in some respects. I have made an image of the working system as-is, with both auto-mount and auto-login disabled, and will make it available to @shanti-bhardwa for testing. I am sorting out my Library options fixing any that don't work and downloading a huge 40GB Wikihow module to add to our biggest library. Many thanks for your diagnostic efforts guys! |
I'll be waiting for my knighthood from her majesty the queen for this work to expand the commonwealth. ;) |
Arise, Sir Jerry! (Well, I'll support your case to Her Maj anyway). A. |
notes on fuse and gvfs |
That syntax seems to have worked, the partition sda6 now appears as
library on the FS.
Should I go ahead now and install IIAB?
…On Fri, Jun 10, 2022 at 1:42 PM Tim Moody ***@***.***> wrote:
I should add that the bind mount will still work even for an internal
partition if standardization is the top priority.
—
Reply to this email directly, view it on GitHub
<#3247 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AZLTUR3OOIA4W7LR4XPDR4DVOMZ2XANCNFSM5YJZ7ZFA>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
Please! As explained at https://download.iiab.io, IIAB install instructions have been shortened to:
[1] There were significant resilience improvements over the past month: [2] Just FYI in coming days (or coming weeks, depending on Kiwix's decision) kiwix-tools 3.3.1 should released and included as a part of IIAB, somewhat enhancing how ZIM files work (like Wikipedia, etc), potentially also improving search: |
Shanti
Well done for getting that going!
Note there is a small typo in your fstab syntax just in case we cut and
paste it
LABEL=IIAB-LIBRART-EXT
I still have to get over to TT at Pentland and collect another T620 or two
to do a "Final" build for them. So I am using an old laptop with a big disk
for now. Been v. busy with Tim and Jerry on our diagnostics - you can
follow most of it on #3247 if you
haven't already found it.
Regards
Andrew
…On Sat, Jun 11, 2022 at 12:30 PM shanti-bhardwa ***@***.***> wrote:
Done some testing on HP t620 with 16GB flash drive. Results look
promising. Here is a summary.
1. Built fresh Mint 20.3. from .iso file and installed it on t620. I
am using Raspberry PI 2B USB WI-FI dongle as my WI-FI interface.
2. Updated the system using sudo apt update & sudo apt upgrade to
bring everything to current levels. Added Gparted as an addition utility.
3. Created two folders on root with sudo mkdir /library and sudo mkdir
/library-ext
4. Formatted my external drive. This is a 32GB SSD in a SATA/USB caddy
with label IIAB-LIBRARY-EXT with ext4 file format. I used these two commands
sudo mkfs.ext4 /dev/sdb1 and sudo e2label /dev/sdb1 IIAB-LIBRARY-EXT.
can also be done using Disks utility or from File Manager, but have not
tested this.
5. Installed the drive and then created a folder with sudo mkdir
/library-ext/library on this USB device.
6. Then added my /etc/fstab entries as follows. These two entries tell
me that USB will always be mounted at /library-ext by mount at boot time &
bind makes the two directory structures the same. IIAB puts everthing under
/library so at IIAB build time only the /library-ext/library on USB is used
as the real srorage,
LABEL=IIAB-LIBRART-EXT /library-ext ext4 defaults,nofail 0 0
/library-ext/library /library none defaults,bind,nofail, 0 0
7.
8. Install IIAB 8.0. Install went without a hitch. The sudo
iiab-diagnostics is here : http:/sprunge.us/InAh1R
9. Downloaded few modules and worked fine.
10 . Thinks I have noticed. df -h shows the usb drive mounted on
/media/usb0 & in on admin console cannot get system storage.
10. The machine is on vpn at 10.8.0.82. You can ssh user is sb01 and
password is shanti
—
Reply to this email directly, view it on GitHub
<#3247 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AZLTUR7TZSS7EFECJDUOQ7TVOR2DVANCNFSM5YJZ7ZFA>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
Just FYI @andrewclark5 @shanti-bhardwa:
|
|
The device should be partitioned:
|
The original instructions intentionally allowed for
I've reworked the FAQ instructions a bit to try to make that (more) explicitly clear. |
@tim-moody that should work, but you could do '/library-ext /library bind' and not have the extra level. Think the 'copy function' would just see the unified underlying filesystem and the second target drive would still be mounted in /media/. Once /library is populated on the 'master data drive' mounted via fstab one could just copy the contents to a new target drive, then redeploy on a second machine with just the fstab entries without the need for symlinks. What needs to be on the cloned target is the library.xml from the donor to match what is on the '/library partition' that would be #3052, to save manual intervention so library.xml travels with the data in /library.
Originally posted by @jvonau in #3106 (comment)
@jvonau @tim-moody @shanti-bhardwa
I'd be grateful for final guidance on how best to configure systems for easy deployment bearing the following in mind:-
a. library.xml in /library/zims
b. iiab-library.xml in /opt/iiab/iiab-admin-console/assets
( as well as a deprecated one which I assume I can ignore.)
a. is no problem as it will be copied with /library.
b. would be more problematic...
The text was updated successfully, but these errors were encountered: