Agent don't collect all information from remote host

[smurdza]

Bug reporting acknowledgment

Yes, I read it

Professional support

None

Describe the bug

When I do remote inventory on the host:

Bellow information from CLI command: dmidecode

Manufacturer: Huawei
        Product Name: TaiShan 2280 V2
        Version: To be filled by O.E.M.
        Serial Number: 2102312XXXXXXXXX
        UUID: 441990c0-XXXX-XXXX-ea11-fd65deXXXXX
        Wake-up Type: Power Switch
        SKU Number: To be filled by O.E.M.
        Family: To be filled by O.E.M.

Agent don't collect information from remote host about UUID and SN, and don't send it to GLPI.
I check xml files and this inventory file don't have this information.

OS on remote host information:
Operating System: EulerOS 2.0 (SP10)
Kernel: Linux 4.19.90-vhulk2207.2.0.h1231.eulerosv2r10.aarch64
Architecture: arm64

To reproduce

1. Configure remote inventory on master host ( OS RockyLinux 9.5 / GLPI-Agent 1.11-2)
2. Add remote host to remote list EulerOS 2.0 (SP10)
3. Start glpi-agent on master host
4. Information about remote host UUID and SN don't send to GLPI

Expected behavior

Agent collect information from remote host about UUID / SN and send it to GLPI.

Operating system

Linux

GLPI Agent version

v1.11

GLPI version

10.0.18

GLPIInventory plugin or other plugin version

Not applicable

Additional context

No response

[g-bougard]

Hi @smurdza

can you confirm you're using a user with administrative privileges for the remote access ?

As the remote computer seems to be arm64 arch, maybe we are missing something. Can you share the output of the command when running with --debug --debug options to enable debug2 level ? Actually the inventory itself is not required, only the RemoteIventory task run log.

About dmidecode, can you confirm it is installed on the remote computer and what is its full path on the system ?

[smurdza]

1. Yes, I am using administrator code.
2. Yes it is arm64
3. From debug 2 log:
   It is part hardware from whole log...
   ---> REMOTE HOST (arm64)
   "hardware": {
   "chassis_type": "Main Server Chassis",
   "datelastloggeduser": "Thu Mar 20 10:37",
   "defaultgateway": "11.111.111.209",
   "lastloggeduser": "adm",
   "memory": 515201,
   "name": "SERV-1-WAW",
   "swap": 32767,
   "vmsystem": "Physical"
   },
   ----> MASTER HOST (x86)
   "hardware": {
   "chassis_type": "Other",
   "datelastloggeduser": "Thu Mar 20 10:33",
   "defaultgateway": "10.111.111.11",
   "dns": "1x2.xx.xxx.253/1x2.xxx.xxx.25x",
   "lastloggeduser": "adm",
   "memory": 7697,
   "name": "master",
   "swap": 8091,
   "uuid": "xxe6xxxx-35f2-fd6d-xxx-xxxx0xxx90xx",
   "vmsystem": "VMware",
   "workgroup": "serwer.pl"
4. Log from debug 1

glpi-agent-debug.txt

5.Path to dmidecode on remote serwer: /sbin/dmidecode

[g-bougard]

Hi @smurdza

I requested debug2 level log (using --debug option 2 times or setting debug = 2 in config), but it still helps to understand what could have happened:

1. GLPI::Agent::Task::Inventory::Generic::Dmidecode is not used. This may have 2 reasons:
   1. dmidecode is not in the PATH, meaning here /sbin is not in used user's PATH. That's a point you can check.
   2. dmidecode output parsing fails. In debug2 level, I would have been able to tell you if that's the problem. Then, here I need dmidecode output (unless you find it is not in the PATH and fixing this fixes the inventory).
2. GLPI::Agent::Task::Inventory::Linux::ARM::Board module is started. This module can set SSN analyzing /proc/cpuinfo file content. Can you share this output ?

[smurdza]

Hi @g-bougard

1. User can use the PATH with sudo.
   without SUDO:

~]$ dmidecode
# dmidecode 3.2
/sys/firmware/dmi/tables/smbios_entry_point: Permission denied
Scanning /dev/mem for entry point.
/dev/mem: Permission denied

with SUDO

~]$ sudo dmidecode
# dmidecode 3.2
Getting SMBIOS data from sysfs.
SMBIOS 3.2.0 present.
Table at 0x2F510000.
 
Handle 0x0000, DMI type 0, 26 bytes
BIOS Information
        Vendor: Huawei Corp.
        Version: 1.79
        Release Date: 08/21/2021
        Address: 0x5F000
        Runtime Size: 644 kB
        ROM Size: 6144 kB
        Characteristics:
                PCI is supported
                BIOS is upgradeable
                BIOS shadowing is allowed
                Boot from CD is supported
                Selectable boot is supported
                EDD is supported

                Japanese floppy for NEC 9800 1.2 MB is supported (int 13h)
                Japanese floppy for Toshiba 1.2 MB is supported (int 13h)
                5.25"/360 kB floppy services are supported (int 13h)
                5.25"/1.2 MB floppy services are supported (int 13h)
                3.5"/720 kB floppy services are supported (int 13h)
                3.5"/2.88 MB floppy services are supported (int 13h)
                8042 keyboard services are supported (int 9h)
                CGA/mono video services are supported (int 10h)
                ACPI is supported
                USB legacy is supported
                BIOS boot specification is supported
                Targeted content distribution is supported
                UEFI is supported
        BIOS Revision: 1.79

Demicode Agent Process log:

[Thu Mar 20 10:34:30 2025][debug] Running GLPI::Agent::Task::Inventory::Generic::Dmidecode
[Thu Mar 20 10:34:30 2025][debug] Running GLPI::Agent::Task::Inventory::Generic::Dmidecode::Battery
[Thu Mar 20 10:34:30 2025][debug2] executing dmidecode
[Thu Mar 20 10:34:30 2025][debug] Running GLPI::Agent::Task::Inventory::Generic::Dmidecode::Bios
[Thu Mar 20 10:34:30 2025][debug2] executing dmidecode
[Thu Mar 20 10:34:30 2025][debug] Running GLPI::Agent::Task::Inventory::Generic::Dmidecode::Hardware
[Thu Mar 20 10:34:30 2025][debug2] executing dmidecode
[Thu Mar 20 10:34:30 2025][debug] Running GLPI::Agent::Task::Inventory::Linux
[Thu Mar 20 10:34:30 2025][debug] Running GLPI::Agent::Task::Inventory::Linux::Hardware
[Thu Mar 20 10:34:30 2025][debug] Running GLPI::Agent::Task::Inventory::Vmsystem
[Thu Mar 20 10:34:30 2025][debug2] executing /sbin/sysctl -n security.jail.jailed
[Thu Mar 20 10:34:30 2025][debug] Running GLPI::Agent::Task::Inventory::Linux::Memory
[Thu Mar 20 10:34:30 2025][debug] Running GLPI::Agent::Task::Inventory::Generic::Dmidecode::Memory
[Thu Mar 20 10:34:30 2025][debug2] executing dmidecode
[Thu Mar 20 10:34:30 2025][debug] Running GLPI::Agent::Task::Inventory::Generic::Dmidecode::Ports
[Thu Mar 20 10:34:30 2025][debug2] executing dmidecode
[Thu Mar 20 10:34:30 2025][debug] Running GLPI::Agent::Task::Inventory::Generic::Dmidecode::Psu
[Thu Mar 20 10:34:30 2025][debug2] executing dmidecode
[Thu Mar 20 10:34:30 2025][debug] Running GLPI::Agent::Task::Inventory::Generic::Dmidecode::Slots
[Thu Mar 20 10:34:30 2025][debug2] executing dmidecode

[Thu Mar 20 10:38:00 2025][debug2] [worker 2] SERWER1-2025-03-18-12-38-54,   GLPI::Agent::Task::Inventory::Generic::Dmidecode::Battery disabled: implicit dependency GLPI::Agent::Task::Inventory::Generic::Dmidecode not enabled
[Thu Mar 20 10:38:00 2025][debug2] [worker 2] SERWER1-2025-03-18-12-38-54,   GLPI::Agent::Task::Inventory::Generic::Dmidecode::Bios disabled: implicit dependency GLPI::Agent::Task::Inventory::Generic::Dmidecode not enabled
[Thu Mar 20 10:38:00 2025][debug2] [worker 2] SERWER1-2025-03-18-12-38-54,   GLPI::Agent::Task::Inventory::Generic::Dmidecode::Hardware disabled: implicit dependency GLPI::Agent::Task::Inventory::Generic::Dmidecode not enabled
[Thu Mar 20 10:38:00 2025][debug2] [worker 2] SERWER1-2025-03-18-12-38-54,   GLPI::Agent::Task::Inventory::Generic::Dmidecode::Memory disabled: implicit dependency GLPI::Agent::Task::Inventory::Generic::Dmidecode not enabled
[Thu Mar 20 10:38:00 2025][debug2] [worker 2] SERWER1-2025-03-18-12-38-54,   GLPI::Agent::Task::Inventory::Generic::Dmidecode::Ports disabled: implicit dependency GLPI::Agent::Task::Inventory::Generic::Dmidecode not enabled
[Thu Mar 20 10:38:00 2025][debug2] [worker 2] SERWER1-2025-03-18-12-38-54,   GLPI::Agent::Task::Inventory::Generic::Dmidecode::Psu disabled: implicit dependency GLPI::Agent::Task::Inventory::Generic::Dmidecode not enabled
[Thu Mar 20 10:38:00 2025][debug2] [worker 2] SERWER1-2025-03-18-12-38-54,   GLPI::Agent::Task::Inventory::Generic::Dmidecode::Slots disabled: implicit dependency GLPI::Agent::Task::Inventory::Generic::Dmidecode not enabled

GLPI::Agent::Task::Inventory::Linux::ARM Agent Process log:

[Thu Mar 20 10:34:21 2025][debug2]   GLPI::Agent::Task::Inventory::Linux::ARM::Board disabled: implicit dependency GLPI::Agent::Task::Inventory::Linux::ARM not enabled
[Thu Mar 20 10:34:21 2025][debug2]   GLPI::Agent::Task::Inventory::Linux::ARM::CPU disabled: implicit dependency GLPI::Agent::Task::Inventory::Linux::ARM not enabled
[Thu Mar 20 10:34:21 2025][debug2] module GLPI::Agent::Task::Inventory::Linux::Alpha disabled
[Thu Mar 20 10:34:21 2025][debug2]   GLPI::Agent::Task::Inventory::Linux::Alpha::CPU disabled: implicit dependency GLPI::Agent::Task::Inventory::Linux::Alpha not enabled

[Thu Mar 20 10:37:08 2025][debug] [worker 1] SERWER1-2025-03-18-12-28-45, Running GLPI::Agent::Task::Inventory::Linux::ARM
[Thu Mar 20 10:37:08 2025][debug] [worker 1] SERWER1-2025-03-18-12-28-45, Running GLPI::Agent::Task::Inventory::Linux::ARM::Board
[Thu Mar 20 10:37:08 2025][debug2] [worker 1] SERWER1-2025-03-18-12-28-45, [libssh2] Trying to read '/proc/cpuinfo' via sftp subsystem
[Thu Mar 20 10:37:08 2025][debug] [worker 1] SERWER1-2025-03-18-12-28-45, Running GLPI::Agent::Task::Inventory::Linux::ARM::CPU
[Thu Mar 20 10:37:08 2025][debug2] [worker 1] SERWER1-2025-03-18-12-28-45, [libssh2] Trying to read '/proc/cpuinfo' via sftp subsystem

[g-bougard]

Hi @smurdza

in the output you provided before, I only see GLPI::Agent::Task::Inventory::Generic::Dmidecode used for the local inventory, not during the RemoteInventory task. Do you mean you changed something in the remote access so now the module is enabled ?

You're talking about sudo, do you mean you're using a remote user which can only run dmidecode with sudo ?

Is the dmidecode output you shared complete ? I don't see any S/N or UUID in that output. Dmidecode is intended to produce more detailed datas, at least on intel arch, but I really don't know if it can do on arm arch.

I also requested /proc/cpuinfo file content but you didn't share it. I think this is the more important point to enhance Arm based computers inventory.

[smurdza]

Hi, @g-bougard
Sorry for my answer, but i can't share you full debug2.

/// Do you mean you changed something in the remote access so now the module is enabled ?
No.

[Thu Mar 20 10:34:36 2025][debug] Starting remoteinventory worker[1] to handle ssh://remote@10.111.111.111
[Thu Mar 20 10:34:36 2025][debug2] Using libssh2 1.9.0 for ssh remote
[Thu Mar 20 10:34:36 2025][debug2] [libssh2] Connecting to '10.111.111.111' remote host...
[Thu Mar 20 10:34:36 2025][debug2] [libssh2] Check remote host key...
[Thu Mar 20 10:34:36 2025][debug2] [libssh2] Trying publickey authentication using id_rsa key...
[Thu Mar 20 10:34:36 2025][debug2] [libssh2] Authenticated on 10.111.111.111 remote with id_rsa key
[Thu Mar 20 10:34:37 2025][debug2] [libssh2] Running "id -u"...
[Thu Mar 20 10:34:37 2025][warning] You should execute remote inventory as super-user on remote host
[Thu Mar 20 10:34:37 2025][debug2] [libssh2] Trying to read '.glpi-agent-deviceid' via sftp subsystem
[Thu Mar 20 10:34:37 2025][info] [worker 1] SERWER1-2025-03-18-12-28-45, New inventory from SERWER1-2025-03-18-12-28-45 for server0 (tag=Network)
[Thu Mar 20 10:34:37 2025][debug2] [worker 1] SERWER1-2025-03-18-12-28-45, [libssh2] Running "uname -s"...


[Thu Mar 20 10:34:37 2025][debug2] [worker 1] SERWER1-2025-03-18-12-28-45, Testing "which dmidecode >/dev/null"...
[Thu Mar 20 10:34:37 2025][debug2] [worker 1] SERWER1-2025-03-18-12-28-45, [libssh2] Running "dmidecode"...
[Thu Mar 20 10:34:37 2025][debug2] [worker 1] SERWER1-2025-03-18-12-28-45, module GLPI::Agent::Task::Inventory::Generic::Dmidecode disabled
[Thu Mar 20 10:34:37 2025][debug2] [worker 1] SERWER1-2025-03-18-12-28-45,   GLPI::Agent::Task::Inventory::Generic::Dmidecode::Battery disabled: implicit dependency GLPI::Agent::Task::Inventory::Generic::Dmidecode not enabled
[Thu Mar 20 10:34:37 2025][debug2] [worker 1] SERWER1-2025-03-18-12-28-45,   GLPI::Agent::Task::Inventory::Generic::Dmidecode::Bios disabled: implicit dependency GLPI::Agent::Task::Inventory::Generic::Dmidecode not enabled
[Thu Mar 20 10:34:37 2025][debug2] [worker 1] SERWER1-2025-03-18-12-28-45,   GLPI::Agent::Task::Inventory::Generic::Dmidecode::Hardware disabled: implicit dependency GLPI::Agent::Task::Inventory::Generic::Dmidecode not enabled
[Thu Mar 20 10:34:37 2025][debug2] [worker 1] SERWER1-2025-03-18-12-28-45,   GLPI::Agent::Task::Inventory::Generic::Dmidecode::Memory disabled: implicit dependency GLPI::Agent::Task::Inventory::Generic::Dmidecode not enabled
[Thu Mar 20 10:34:37 2025][debug2] [worker 1] SERWER1-2025-03-18-12-28-45,   GLPI::Agent::Task::Inventory::Generic::Dmidecode::Ports disabled: implicit dependency GLPI::Agent::Task::Inventory::Generic::Dmidecode not enabled
[Thu Mar 20 10:34:37 2025][debug2] [worker 1] SERWER1-2025-03-18-12-28-45,   GLPI::Agent::Task::Inventory::Generic::Dmidecode::Psu disabled: implicit dependency GLPI::Agent::Task::Inventory::Generic::Dmidecode not enabled
[Thu Mar 20 10:34:37 2025][debug2] [worker 1] SERWER1-2025-03-18-12-28-45,   GLPI::Agent::Task::Inventory::Generic::Dmidecode::Slots disabled: implicit dependency GLPI::Agent::Task::Inventory::Generic::Dmidecode not enabled

/// You're talking about sudo, do you mean you're using a remote user which can only run dmidecode with sudo ?---
Yes, it is. Only sudo mode can read.

/// I don't see any S/N or UUID in that output
Because, agent don,t read it.

// /proc/cpuinfo

processor       : 0
BogoMIPS        : 200.00
cpu MHz         : 2600.000
Features        : fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma dcpop asimddp asimdfhm
CPU implementer : 0x48
CPU architecture: 8
CPU variant     : 0x1
CPU part        : 0xd01
CPU revision    : 0
 
processor       : 1
BogoMIPS        : 200.00
cpu MHz         : 2600.000
Features        : fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma dcpop asimddp asimdfhm
CPU implementer : 0x48
CPU architecture: 8
CPU variant     : 0x1
CPU part        : 0xd01
CPU revision    : 0
 
processor       : 2
BogoMIPS        : 200.00
cpu MHz         : 2600.000
Features        : fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma dcpop asimddp asimdfhm
CPU implementer : 0x48
CPU architecture: 8
CPU variant     : 0x1
CPU part        : 0xd01
CPU revision    : 0
 
processor       : 3
BogoMIPS        : 200.00
cpu MHz         : 2600.000
Features        : fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma dcpop asimddp asimdfhm
CPU implementer : 0x48
CPU architecture: 8
CPU variant     : 0x1
CPU part        : 0xd01
CPU revision    : 0

and processor : 128 report the same.

[g-bougard]

/// You're talking about sudo, do you mean you're using a remote user which can only run dmidecode with sudo ?---
Yes, it is. Only sudo mode can read.

It's better to use an admin user in the context of RemoteInventory task. If you don't, you can loose some critical data like you here. If dmidecode provides the missing data when run with sudo (and I'm still not convinced it does as you didn't provide an output showing it can do), you still can create a script named in dmidecode in the user PATH environment which will run dmidecode with sudo:

1. Check path known in that user context: ssh remotehost sh -c 'echo $PATH'
2. if ~/bin is in the path and before /sbin, create an executable ~/bin/dmidecode file with

#! /bin/sh
exec sudo /sbin/dmidecode $*

/// I don't see any S/N or UUID in that output
Because, agent don,t read it.

I was meaning in dmidecode command output you provided. My problem is you shared an output which is not coherent with what you reported in issue description.

// /proc/cpuinfo

Nothing seems useful from this output.

[smurdza]

Hi, @g-bougard

I was meaning in dmidecode command output you provided. My problem is you shared an output which is not coherent with what you reported in issue description.

It is coherent, because is share you only part of whole dmidecode.
First part of output demi (three firsts section), secound section contens information on SN and UUID:

# dmidecode 3.2
Getting SMBIOS data from sysfs.
SMBIOS 3.2.0 present.
Table at 0x2F510000.
 
Handle 0x0000, DMI type 0, 26 bytes
BIOS Information
        Vendor: Huawei Corp.
        Version: 1.79
        Release Date: 08/21/2021
        Address: 0x5F000
        Runtime Size: 644 kB
        ROM Size: 6144 kB
        Characteristics:
                PCI is supported
                BIOS is upgradeable
                BIOS shadowing is allowed
                Boot from CD is supported
                Selectable boot is supported
                EDD is supported
                Japanese floppy for NEC 9800 1.2 MB is supported (int 13h)
                Japanese floppy for Toshiba 1.2 MB is supported (int 13h)
                5.25"/360 kB floppy services are supported (int 13h)
                5.25"/1.2 MB floppy services are supported (int 13h)
                3.5"/720 kB floppy services are supported (int 13h)
                3.5"/2.88 MB floppy services are supported (int 13h)
                8042 keyboard services are supported (int 9h)
                CGA/mono video services are supported (int 10h)
                ACPI is supported
                USB legacy is supported
                BIOS boot specification is supported
                Targeted content distribution is supported
                UEFI is supported
        BIOS Revision: 1.79
 
Handle 0x0001, DMI type 1, 27 bytes
System Information
        Manufacturer: Huawei
        Product Name: TaiShan 2280 V2
        Version: To be filled by O.E.M.
        Serial Number: 2102312XXXXXXXXX
        UUID: 441990c0-XXXX-XXXX-ea11-fd65deXXXXX
        Wake-up Type: Power Switch
        SKU Number: To be filled by O.E.M.
        Family: To be filled by O.E.M.
 
Handle 0x0002, DMI type 3, 25 bytes
Chassis Information
        Manufacturer: Huawei
        Type: Main Server Chassis
        Lock: Not Present
        Version: To be filled by O.E.M.
        Serial Number: To be filled by O.E.M.
        Asset Tag: To be filled by O.E.M.
        Boot-up State: Safe
        Power Supply State: Safe

Check path known in that user context: ssh remotehost sh -c 'echo $PATH'
if ~/bin is in the path and before /sbin, create an executable ~/bin/dmidecode file with

ssh user1@10.111.111.111 sh -c 'echo $PATH'
 
Authorized users only. All activities may be monitored and reported.

2. On the user1 accout on server:

$ echo $PATH
/bin:/sbin:/usr/local/sbin:/usr/sbin:/bin:/sbin:/usr/local/sbin:/usr/sbin:/usr/local/bin:/usr/bin:/usr/local/sbin:/usr/sbin:/bin:/sbin:/usr/bin:/usr/sbin:/bin:/sbin:/usr/bin:/usr/sbin

ls -ltr /sbin/dmidecode
-rwxr-xr-x 1 root root 133856 Aug 2 2023 /sbin/dmidecode

[g-bougard]

Strange PATH with a lot of duplicated entry.

Anyway, only /bin seems to be before /sbin. This way, check /bin/dmidecode, then you can create the script I proposed here. And this should work.

About incomplete datas I noticed, you would have to tell you didn't give the full output. Having the full output can help to fix any parsing issue, but if you don't want to share, I can't confirm you dmidecode parsing on arm arch is still correct. So we can only hope current parsing is correct🤞. It should be looking at your last sharing.

So finally, for me, there's no issue glpi-agent side. And this is only a remote access configuration problem. Then I'll convert this issue as Q&A discussion.