From 92e690ca35c8618d10b955c8a49211aa84845161 Mon Sep 17 00:00:00 2001
From: jwrauch <jwrauch@web>
Date: Tue, 24 Jun 2025 18:33:51 +0000
Subject: [PATCH 1/2]

---
 ...firming_my_preferred_content_understanding.mdwn | 14 ++++++++++++++
 1 file changed, 14 insertions(+)
 create mode 100644 doc/forum/Confirming_my_preferred_content_understanding.mdwn

diff --git a/doc/forum/Confirming_my_preferred_content_understanding.mdwn b/doc/forum/Confirming_my_preferred_content_understanding.mdwn
new file mode 100644
index 0000000000..67ca20d911
--- /dev/null
+++ b/doc/forum/Confirming_my_preferred_content_understanding.mdwn
@@ -0,0 +1,14 @@
+I've been using git-annex for a little while now to keep two copies of my data backed up across three and soon to be four external hard drives.  I recently came across the preferred content expressions and realized that they could make my life a lot easier, but I want to make sure I understand them correctly first. 
+
+As I mentioned, I want two copies of my data stored across several external hard drives.  Right now I keep track of this manually and I don't try to balance the data between drives.  From what I understand, I can add the repository from each drive to the archive group.  And then call 
+
+```
+   git annex groupwanted archive (not (copies=archive:2 and balanced=archive:lackingcopies))
+   git annex wanted here groupwanted
+```
+
+And then when I sync the repositories it will try and balance the data between themselves while ensuring there are at least two copies of the data.  I can also easily add hard drives to the system by cloning a repository on them and then adding it to the group.  
+
+Do I have the correct understanding here or am I missing something?  Are there any suggestions or advice you have for this setup?  Is using `fullybalanced` or `sizebalanced` better?  
+
+

From ce99131e8e08846326cceeb5ffc7d9f8499f4bac Mon Sep 17 00:00:00 2001
From: xentac <xentac@web>
Date: Tue, 24 Jun 2025 20:54:16 +0000
Subject: [PATCH 2/2] Added a comment

---
 .../comment_5_068d43f41f5c0f7537b3f7d9e4f3c8ad._comment   | 8 ++++++++
 1 file changed, 8 insertions(+)
 create mode 100644 doc/bugs/adb_fchown_error/comment_5_068d43f41f5c0f7537b3f7d9e4f3c8ad._comment

diff --git a/doc/bugs/adb_fchown_error/comment_5_068d43f41f5c0f7537b3f7d9e4f3c8ad._comment b/doc/bugs/adb_fchown_error/comment_5_068d43f41f5c0f7537b3f7d9e4f3c8ad._comment
new file mode 100644
index 0000000000..235b6f8354
--- /dev/null
+++ b/doc/bugs/adb_fchown_error/comment_5_068d43f41f5c0f7537b3f7d9e4f3c8ad._comment
@@ -0,0 +1,8 @@
+[[!comment format=mdwn
+ username="xentac"
+ avatar="http://cdn.libravatar.org/avatar/773b6c7b0dc34f10b66aa46d2730a5b3"
+ subject="comment 5"
+ date="2025-06-24T20:54:16Z"
+ content="""
+To close the loop (for anyone searching for this later), I had to setgid (`chown g+s <directory`) on the directories created by git annex so that `adb push` could successfully fchown them. It was able to write the contents of the file (the percentage went up and I could see the file being written from an `adb shell` but once adb tried to `fchown` it would fail and delete the file from the filesystem. Now my files are uploading successfully again.
+"""]]