summaryrefslogtreecommitdiff
path: root/open_issues/active_vs_passive_symlink_translator.mdwn
diff options
context:
space:
mode:
Diffstat (limited to 'open_issues/active_vs_passive_symlink_translator.mdwn')
-rw-r--r--open_issues/active_vs_passive_symlink_translator.mdwn44
1 files changed, 0 insertions, 44 deletions
diff --git a/open_issues/active_vs_passive_symlink_translator.mdwn b/open_issues/active_vs_passive_symlink_translator.mdwn
deleted file mode 100644
index cbd9b077..00000000
--- a/open_issues/active_vs_passive_symlink_translator.mdwn
+++ /dev/null
@@ -1,44 +0,0 @@
-[[!meta copyright="Copyright © 2011 Free Software Foundation, Inc."]]
-
-[[!meta license="""[[!toggle id="license" text="GFDL 1.2+"]][[!toggleable
-id="license" text="Permission is granted to copy, distribute and/or modify this
-document under the terms of the GNU Free Documentation License, Version 1.2 or
-any later version published by the Free Software Foundation; with no Invariant
-Sections, no Front-Cover Texts, and no Back-Cover Texts. A copy of the license
-is included in the section entitled [[GNU Free Documentation
-License|/fdl]]."]]"""]]
-
-[[!tag open_issue_documentation open_issue_hurd]]
-
-IRC, freenode, #hurd, 2011-07-25
-
-Set an *active* (not *passive*) `/hurd/symlink` translator on a node.
-
- < antrik> that's strange: the file doesn't look like a symlink in ls output
- -- but it behaves like one...
- < antrik> using firmlink instead of symlink yields less confusing
- results...
- < gg0> how does it behaves like one?
- < antrik> perhaps the symlink mechanism only fully works for a passive
- symlink translator, not an active one
- < antrik> gg0: if you access it, you actually get the linked file contents
- < antrik> it's only ls that's confused
- < antrik> it might be because ls -l uses O_NOFOLLOW, which results in
- O_NOTRANS, so it sees the original file contents
- < gg0> stat says it's still 12264 bytes
- < antrik> stat also seems to use NOFOLLOW
- < antrik> wc will show the "correct" size
- < gg0> ok
- < antrik> if you set it as passive translator, it works as expected... but
- then you better don't forget removing it, as it won't go away after a
- reboot :-)
- < antrik> but as I said, you can just ignore the weirdness -- or use
- firmlink instead
- < antrik> the thing is, if symlink is set as a passive translator, the
- filesystem handles it specially, so it really looks like a symlink to
- programs using NOFOLLOW. that's not the case with an active symlink... so
- programs using NOFOLLOW simply do not see the active symlink at all
- < antrik> firmlink OTOH ignores NOFOLLOW, so you always see the linked-to
- file
-
- * [[hurd/translator/short-circuiting]]