Can list and see files/dirs in Console, but missing dirs/files if listing via SFTP client

September 13, 2019 229 views
Miscellaneous

First off I’m a linux newb, trying to follow instructions found on the ‘net. So bear with me if I’m missing the obvious.

I’m trying to troubleshoot a droplet that won’t load due to a kernel panic.
Screenshot - https://gyazo.com/ef9bb5fe876076b414c0fb0de34a804b

So I can log on using recovery image, mount the disk image and chroot into the system.

Now if I cd into /boot/ I can see a list of images and other files/dir.
If I do the same for the /var/ I can see the www directory listed

BUT if I log on using a SFTP client (SmartFTP & Winscp), using the provided temp root password, I can not see the same list in /boot/, nor can I see the /www dir in /var/.

Opening a terminal window from within those clients also fails to see these directories and/or files.

Screenshot of /boot/ in DO console -https://gyazo.com/4f77c5f1ab2b8f068f076ebf737fbd57

same dir but using sftp client
https://gyazo.com/275949d69a7ee61b23c6ec89cd25820f

/var/ in console
https://gyazo.com/a32a2ddd077cdf147a75ad7d7b26ce22

/var/ from sftp client

https://gyazo.com/f3600697102f4827cb6077b74232655c

So can someone point out what I’m doing wrong here and how to fix the issue of the missing dir/files in sftp.
cheers
Steve

2 comments
  • Hi Alex, No the issue still remains, and I haven’t heard back from DO support as of yet.

    There is no Grub menu that I can see, or if there is, I don’t know how to access it.

    For what it’s worth this droplet uses Grub2, and internal kernel management, which from my reading, means the options are different from pre-Grub2. Am I right?

    I can logon via recovery iso, as I did that to mount & check the file system, but cannot follow the the instructions on how to replace the faulty kernel. It’s very gobbledygook to me. I tried, but the recovery iso terminal just hangs.

    I’d appreciate any step by step help you could offer, as I’ve been offline for over 24hr now. I must say, I’m a little disappointed in Support not responding as of yet.

  • Seems my earlier reply was not approved. So much for DO offering support. Over 24 hrs since lodegement of support ticket and no reply. Meanwhile I’ve several sites offline.

    Anyway, the issue I’m facing after reading and reading all the documentation of changing the kernel is that I get 3 errors which seem to be show stoppers.

    firstly when running “sudo apt-get install linux-image-unsigned-4.18.0-25-generic linux-headers-4.18.0-25-generic” I get the following error

    “update-initramfs: Generating /boot/initrd.img-4.15.0-62-generic
    /sbin/ldconfig.real: Can’t create temporary cache file /etc/ld.so.cache~: No such file or directory”

    And if I persist and then run “ sudo update-grub” I get the error “/usr/sbin/grub-probe: error: failed to get canonical path of `overlay’.”

    All my searching for answers to those errors seems to point to the fact I’m a “fake” root, i.e. chroot, and therefore I don’t have access permissions for those paths/files.

    So I’m stumped for a solution.

    Really DO, it shouldn’t be this hard to be able to select a different kernel.

4 Answers

Hello,

Are you still having issues when trying to start the droplet? Can you choose another kernel to boot from when starting the droplet (in the GRUB menu)?

The other option will be to boot your droplet using Digitalocean’s recovery iso, here are the steps to do it https://www.digitalocean.com/docs/droplets/resources/recovery-iso/

After that, you can mount the file system and choose the option to chroot, so you can switch to centos/ubuntu file system. next option is to switch the kernel of centos/ubuntu to recovery kernel and boot again with the original HDD and remove the kernel and install it again.

Let me know how it goes.

Alex

Hi Alex, No the issue still remains, and I haven’t heard back from DO support as of yet.

There is no Grub menu that I can see, or if there is, I don’t know how to access it.

For what it’s worth this droplet uses Grub2, and internal kernel management, which from my reading, means the options are different from pre-Grub2. Am I right?

I can logon via recovery iso, as I did that to mount & check the file system, but cannot follow the the instructions on how to replace the faulty kernel. It’s very gobbledygook to me. I tried, but the recovery iso terminal just hangs.

I’d appreciate any step by step help you could offer, as I’ve been offline for over 24hr now. I must say, I’m a little disappointed in Support not responding as of yet.

To elaborate, when following the instructions found here https://www.digitalocean.com/docs/droplets/how-to/kernel/use-non-default/

I keep getting the following error about canonical path not found “overlay”

https://gyazo.com/2fb49d83a2f892b0a21b246995489c54

plus the final command given in those instructions “sudo reboot” doesn’t work either.

It’s very difficult if not impossible for a novice like myself to repair a borked droplet when DO own recovery instructions don’t work. And still, over 24 hrs later, no response from Support. Very frustrating

Have another answer? Share your knowledge.