[New: openFATE 314483] Automatically set secondary partitions as 'nofail' in /etc/fstab
Feature added by: Mircea Kitsune (MirceaKitsune) Feature #314483, revision 1 Title: Automatically set secondary partitions as 'nofail' in /etc/fstab openSUSE Distribution: Unconfirmed Priority Requester: Mandatory Requested by: Mircea Kitsune (mirceakitsune) Partner organization: openSUSE.org Description: I had a discussion yesterday about how partitions in fstab disappearing (such as unplugging an internal drive) will cause openSUSE to no longer boot and go to the emergency console, unless the 'nofail' parameter is set. For those who wish to see the full thread: http://forums.opensuse.org/english/get-technical-help-here/install-boot-logi... Being a newbie myself, I found this difficult to solve and understand until someone explained me how fstab works. Until then I thought the system had been compromised beyond repair. Some users might remove drives or change partitions without knowing how fstab works. They would then find themselves stuck in an "emergency console" without any information as to why they are there, conclude the only solution is to reinstall the system, and go through a frustrating experience thinking Linux is too unsafe and unstable. My suggestion is defaulting the 'nofail' parameter for all drives except root and swap (possibly the /home partition if it can be detected) when openSUSE / Partitioner automatically add a drive to fstab. There's no point in failing the boot if a partition that's not needed to boot is missing. I have three NTFS partitions mounted in /windows/[letter]/ and immediately added this flag to them after being informed of it, to avoid potential issues in case one of them breaks or I temporarily disconnect it. Separately, I'm also proposing a message in the emergency console about what brought the user there. If possible, mentioning that in /etc/fstab there's a partition that's no longer accessible. The only information currently shown is "Welcome to the emergency console, please give your root password". If the user forgets what changes they made, they might never know what's broken and what to try fixing... at least if they're not experts and have a console command ready. Test Case: Make sure you have a partition on a secondary drive configured to auto-mount in /etc/fstab (for NTFS that should be /windows/[letter]/). Shut down your computer and unplug that drive. When booting into openSUSE, you will be taken to the recovery console before reaching the logon screen, without any information as to why. If you add the 'nofail' parameter to the missing partition in /etc/fstab, it will boot as normal and just ignore the missing drive. Business case (Partner benefit): openSUSE.org: Newbies might not know all the settings in fstab or edit them manually for drives auto-mounted by openSUSE. If a secondary drive breaks or is disconnected, they can be left with an irrecoverable system without even knowing why. Even an expert will need to boot using a Live-CD distro to edit /etc/fstab manually to get the system booting again without that drive. Automatically setting alternative drives as 'nofail' would prevent such risks, and also make most sense in my opinion. -- openSUSE Feature: https://features.opensuse.org/314483
Feature changed by: Carlos Robinson (robin_listas) Feature #314483, revision 3 Title: Automatically set secondary partitions as 'nofail' in /etc/fstab openSUSE Distribution: Unconfirmed Priority Requester: Mandatory Requested by: Mircea Kitsune (mirceakitsune) Partner organization: openSUSE.org Description: I had a discussion yesterday about how partitions in fstab disappearing (such as unplugging an internal drive) will cause openSUSE to no longer boot and go to the emergency console, unless the 'nofail' parameter is set. For those who wish to see the full thread: http://forums.opensuse.org/english/get-technical-help-here/install-boot-logi... Being a newbie myself, I found this difficult to solve and understand until someone explained me how fstab works. Until then I thought the system had been compromised beyond repair. Some users might remove drives or change partitions without knowing how fstab works. They would then find themselves stuck in an "emergency console" without any information as to why they are there, conclude the only solution is to reinstall the system, and go through a frustrating experience thinking Linux is too unsafe and unstable. My suggestion is defaulting the 'nofail' parameter for all drives except root and swap (possibly the /home partition if it can be detected) when openSUSE / Partitioner automatically add a drive to fstab. There's no point in failing the boot if a partition that's not needed to boot is missing. I have three NTFS partitions mounted in /windows/[letter]/ and immediately added this flag to them after being informed of it, to avoid potential issues in case one of them breaks or I temporarily disconnect it. Separately, I'm also proposing a message in the emergency console about what brought the user there. If possible, mentioning that in /etc/fstab there's a partition that's no longer accessible. The only information currently shown is "Welcome to the emergency console, please give your root password". If the user forgets what changes they made, they might never know what's broken and what to try fixing... at least if they're not experts and have a console command ready. Test Case: Make sure you have a partition on a secondary drive configured to auto- mount in /etc/fstab (for NTFS that should be /windows/[letter]/). Shut down your computer and unplug that drive. When booting into openSUSE, you will be taken to the recovery console before reaching the logon screen, without any information as to why. If you add the 'nofail' parameter to the missing partition in /etc/fstab, it will boot as normal and just ignore the missing drive. Business case (Partner benefit): openSUSE.org: Newbies might not know all the settings in fstab or edit them manually for drives auto-mounted by openSUSE. If a secondary drive breaks or is disconnected, they can be left with an irrecoverable system without even knowing why. Even an expert will need to boot using a Live-CD distro to edit /etc/fstab manually to get the system booting again without that drive. Automatically setting alternative drives as 'nofail' would prevent such risks, and also make most sense in my opinion. + Discussion: + #1: Carlos Robinson (robin_listas) (2012-09-28 16:27:09) + No, it can not be the default option. First, only the admin knows which + partitions are optional. Second, if a partition fails to mount for some + reason, it will do so silently, without any error message. If someone, + or a script, tries to use that partition, the writes will go to the + parent partition where it should be mounted. + For example, suppose you mount "/backup" with nofail, and it is a 2 TiB + drive. It fails to mount, then the automated backup script triggers and + does a backup to "/backup", which writes into "/" instead which is a + partition of 20 GiB, fills it fast, and crashes the system (which will + not boot if you try). -- openSUSE Feature: https://features.opensuse.org/314483
Feature changed by: Mircea Kitsune (MirceaKitsune) Feature #314483, revision 4 Title: Automatically set secondary partitions as 'nofail' in /etc/fstab openSUSE Distribution: Unconfirmed Priority Requester: Mandatory Requested by: Mircea Kitsune (mirceakitsune) Partner organization: openSUSE.org Description: I had a discussion yesterday about how partitions in fstab disappearing (such as unplugging an internal drive) will cause openSUSE to no longer boot and go to the emergency console, unless the 'nofail' parameter is set. For those who wish to see the full thread: http://forums.opensuse.org/english/get-technical-help-here/install-boot-logi... Being a newbie myself, I found this difficult to solve and understand until someone explained me how fstab works. Until then I thought the system had been compromised beyond repair. Some users might remove drives or change partitions without knowing how fstab works. They would then find themselves stuck in an "emergency console" without any information as to why they are there, conclude the only solution is to reinstall the system, and go through a frustrating experience thinking Linux is too unsafe and unstable. My suggestion is defaulting the 'nofail' parameter for all drives except root and swap (possibly the /home partition if it can be detected) when openSUSE / Partitioner automatically add a drive to fstab. There's no point in failing the boot if a partition that's not needed to boot is missing. I have three NTFS partitions mounted in /windows/[letter]/ and immediately added this flag to them after being informed of it, to avoid potential issues in case one of them breaks or I temporarily disconnect it. Separately, I'm also proposing a message in the emergency console about what brought the user there. If possible, mentioning that in /etc/fstab there's a partition that's no longer accessible. The only information currently shown is "Welcome to the emergency console, please give your root password". If the user forgets what changes they made, they might never know what's broken and what to try fixing... at least if they're not experts and have a console command ready. Test Case: Make sure you have a partition on a secondary drive configured to auto- mount in /etc/fstab (for NTFS that should be /windows/[letter]/). Shut down your computer and unplug that drive. When booting into openSUSE, you will be taken to the recovery console before reaching the logon screen, without any information as to why. If you add the 'nofail' parameter to the missing partition in /etc/fstab, it will boot as normal and just ignore the missing drive. Business case (Partner benefit): openSUSE.org: Newbies might not know all the settings in fstab or edit them manually for drives auto-mounted by openSUSE. If a secondary drive breaks or is disconnected, they can be left with an irrecoverable system without even knowing why. Even an expert will need to boot using a Live-CD distro to edit /etc/fstab manually to get the system booting again without that drive. Automatically setting alternative drives as 'nofail' would prevent such risks, and also make most sense in my opinion. Discussion: #1: Carlos Robinson (robin_listas) (2012-09-28 16:27:09) No, it can not be the default option. First, only the admin knows which partitions are optional. Second, if a partition fails to mount for some reason, it will do so silently, without any error message. If someone, or a script, tries to use that partition, the writes will go to the parent partition where it should be mounted. For example, suppose you mount "/backup" with nofail, and it is a 2 TiB drive. It fails to mount, then the automated backup script triggers and does a backup to "/backup", which writes into "/" instead which is a partition of 20 GiB, fills it fast, and crashes the system (which will not boot if you try). + #2: Mircea Kitsune (mirceakitsune) (2012-09-29 11:02:40) (reply to #1) + I understand. I'm still new so users with more experience know better + what else might happen. But personally, I'd still suggest an + alternative in some form. I didn't think about what would happen if the + mount folder disappears... assumed any backup / copy tool would notice + it's not there and abort. But if /windows/D is the mounting point of a + partition, and that partition is not mounted, a tool normally backing + up to that might treat /windows/D as a folder and save there. + Still: Running into system issues due to a drive path being + misinterpreted might be more acceptable than the system not booting at + all (with the appearance that it's been permanently compromised). If + you can still boot, you can fix such a problem once you are aware of + it. Not sure if openSUSE could either block the path to a broken + mounting point, or simply inform the user at logon about this problem + and existing dangers. -- openSUSE Feature: https://features.opensuse.org/314483
participants (1)
-
fate_noreply＠suse.de