Score:0

How can Xubuntu read/write to an unmounted HDD

gt flag

This is a follow-on from my question a few days back: How can I stop constant HD journaling?

Setup is now:

  • Windows 10 Pro on SSD
  • Xubuntu 22.04 on SSD: partitioned as 1. /root (ext4) 2. /home (ext4) 3. /swap
  • One internal SATA 2Tb HDD drive partitioned into two, both NTFS
  • Boot is GPT

Problem/issue: Running Windows I can access the Windows SSD partition, internal HDD and any USB-connected external hard drives or USB drives. Running Xubuntu I can access the SSD partitions and I can mount the internal HDD partitions and external drives are automatically mounted. I am transferring backed-up data from external HDs to the internal HDD using Thunar or Dolphin. Transfers are executed without issue. However, a short time later the HDD is subject to continual read or write access. I cannot identify what process this is but it seems to me to be indexing of some sort because eventually it stops.

The problem occurs when I unmount the HDD but the read/write disk access continues. If I shut down Xubuntu then restart the read/write access starts again, still with the HDD not showing as mounted. How can this happen?

Possibly associated with this is an error I received on one shutdown while access was occurring began: GDBus.Error: freedesktop.DBus.Error.InvalidArgs: Type of message... then it disappeared.

Main query is the accessing a drive that it not mounted. Does anyone have an idea of how this is possible? Is it a problem, aside from annoyance? Customer support from the computer manufacturer admitted they knew very little, 'nothing actually', about Linux.

RonJohn avatar
cn flag
"but the read/write disk access continues." Drives have cache, too, that's independent of OS control.
NickN avatar
gt flag
Yes, true, but can such a cache survive a power-off shutdown? The drive activity seems very much to be indexing of some sort because it has now stopped, and no new files added to start it up again. Interestingly, the 'indexing' also occurs in Windows, and has also stopped. Do Seagate drives have an in-built indexing process? Or has the indexing stopped because I turned on System Protection on the HDD in Windows? Basically, I'm still in the dark on this.
RonJohn avatar
cn flag
You say that you think the HDD is indexing something. What evidence do you have for that?
NickN avatar
gt flag
Evidence? Bear in mind that I am a regular user of Linux/Xubuntu but my technical knowledge is limited. What I am hearing is exclusively noise from the HDD; it is not and cannot be noise from the SSD. The noise eventually stops - as above, I have not added new files so no new 'indexing' to carry out. Likewise, the same noise has stopped when running Windows. One small matter to throw into the mix is that I have found that I need to turn off Fast Start in Windows because with it enabled my external HDD (the one with the noise) tends to open in read-only mode; could this be a part of the issue?
RonJohn avatar
cn flag
You keep calling it "indexing" with no justification. The only thing that can happen to dismounted disks is for it to finish writing it's internal buffers.
Score:0
gt flag

I think I may have an answer to my own question. It was becoming more and more apparent that this was a hardware issue, not a Linux/Xubuntu one: the noises effect both Xubuntu and Windows.

Answer: According to this thread https://forums.tomshardware.com/threads/seagate-barracuda-weird-sounds.3522871/ the drive noises I am hearing fit the explanation as to how an SMR (Shingled magnetic recording) disk works. My disk model is an SMR disk - https://www.seagate.com/gb/en/products/cmr-smr-list/ - so RonJohn's comments about caching and buffering are on the money.

Solution: Frequently accessed files stay on the SSD, then fit a smaller 1TB drive to handle occasionally accessed files and keep the 2Tb Seagate for backup only; and, where possible, the 2Tb set to power down when not in use.

I sit in a Tesla and translated this thread with Ai:

mangohost

Post an answer

Most people don’t grasp that asking a lot of questions unlocks learning and improves interpersonal bonding. In Alison’s studies, for example, though people could accurately recall how many questions had been asked in their conversations, they didn’t intuit the link between questions and liking. Across four studies, in which participants were engaged in conversations themselves or read transcripts of others’ conversations, people tended not to realize that question asking would influence—or had influenced—the level of amity between the conversationalists.