jra_samba

13 ms·

91.
by jra_samba 5y ago
Linus had no part in the drafting of GPLv3, so no I'm not surprised by that video at all. I participated in the GPLv3 drafting, and your claim it was full of "one sketchy / dicy deal after another" is NOT correct. You
92.
by jra_samba 5y ago
No, Samba does not need to do inode to file lookups. There are no "open by inode" primitives in SMB1/2/3.
93.
by jra_samba 5y ago
> As GPLv3 was getting developed (in secret largely) The above is incorrect. GPLv3 was developed via a completely open process of development, documented very well here: https://opensource.com/article/18/6/
94.
by jra_samba 5y ago
How are you using Samba -> ceph ? Are you re-exporting a fuse mount or are you using Sambas built-in vfs_ceph module ? The vfs_ceph module should be faster as it's essentially a protocol translator from SMB1/2/3 -> ceph
95.
by jra_samba 5y ago
Sorry Apple fans, but you have been living in the very definition of "The Hotel California". Apple has altered the deal. Pray they do not alter it any further. Now you have to live with the consequences of convenience.
96.
Endless is launching a program to increase PC access equity in the USA(endlessos.org)
1 points by jra_samba 5y ago|0 comments
97.
by jra_samba 5y ago
No so much harlinks, but symlinks are a blight on the POSIX filesystem design. They have caused endless pain and suffering and so many, many CVE's. They need to be eliminated.
98.
by jra_samba 5y ago
Samba actually ran into this with io_uring. It can return less that the client asked for for read/write. We were returning this to the client and the clients... didn't cope well (even though the protocol is supposed to allow this
99.
by jra_samba 5y ago
We have an internal Samba VFS module that can do the same thing for SMB testing. It's really useful !
100.
by jra_samba 5y ago
Thanks for the recommendation. We work pretty closely with Apple's SMB2 client developers to make sure everything works well together with Samba. They're great fun to work with !
101.
by jra_samba 5y ago
I haven't checked, but this may be due to the restriction in CCM/GCM. Here is the relevant comment/code from Samba. * CCM and GCM algorithms must never have their * nonce wrap, or the security of the whole * commun
102.
by jra_samba 5y ago
io_uring has been a game-changer for Samba IO speed. Check out Stefan Metzmacher's talk at SambaXP 2021 (online event) for details: https://www.youtube.com/watch?v=eYxp8yJHpik
103.
by jra_samba 5y ago
At SambaXP last week, Metze showed Samba + io_uring on Linux saturating 100GB ethernet. https://www.youtube.com/watch?v=eYxp8yJHpik Not too shabby !
104.
by jra_samba 5y ago
Sneakers works great for the era, if you just imagine the "magic box" as a fast DES cracker. It even makes the comment at the end about how "it only works on US crypto, not Russian" make sense !
105.
by jra_samba 5y ago
You're talking about "Arrival", not "The Arrival", which is a 1996 movie staring Charlie Sheen about an alien invasion. https://www.imdb.com/title/tt0115571/ Strangely enough the Charlie S
106.
by jra_samba 6y ago
Yes, this is true. Google corporate culture is one of the big reasons I work here. There are few places that take Open Source as seriously as Google does.
107.
by jra_samba 6y ago
Ted, you have to remember why Chromebooks have a way of turning off signed boot. It's actually to comply with GPLv3 requirements for the GPLv3 code used in ChromeOS. Google takes licensing seriously (as both know you and I know :-).
108.
by jra_samba 6y ago
This is a very silly question. Open Source is a development methodology, not a business. It's like saying "There's still no viable business model for science." Just makes no sense. I've been making a living on Open
109.
by jra_samba 6y ago
I got to have lunch with Neil Gaiman, when he visited Google. Got to sit next to him and geek out as a fanboy :-). I also discussed "Lord of Light" with him (my favorite novel), but I'd imagine he had already read it way befo
110.
by jra_samba 6y ago
As this is a protocol level vulnerability, older versions of Samba were also affected. Security release announcement is here: https://www.samba.org/samba/security/CVE-2020-1472.html After a previous vulnerability
111.
by jra_samba 6y ago
> How would one go about implementing case-insensitive path lookup against a SMB share containing a > few million files from userspace? It must be in the kernel. SMB in the kernel is a rather dangerous game. Not saying it can't b
112.
by jra_samba 6y ago
You are incorrect. It is the only way for a user-space application to provide Windows-style case insensitivity. Windows slient sends a name "foo" over SMB2+. Samba does a stat("foo", &st) call, gets ENOENT. Now, does
113.
by jra_samba 6y ago
Yep, that's true. It's the cache misses that kill performance. If the client asks for file "Foo", and the (l)stat fails to find it, then we have to scan the whole directory looking for any case-differing versions of &quo
114.
by jra_samba 6y ago
It's only an extra round trip in the case of an unexpected EOF. File size is returned from SMB2_CREATE and so given the default of a RHW lease then (a) the lease can't be violated - if it is, then all bets are off as the server le
115.
by jra_samba 6y ago
No the client can't assume that. Consider pipelining reads. You can asynchronously send 10 1MB reads. The server can return the data in any order. So the read sent at offset 0 could return last after the server has already returned 9MB
116.
by jra_samba 6y ago
It is a Windows and MacOS bug, but only visible with the io_uring change in behavior. Problem is we have to fix the server, as even when buggy, one billion clients can never be wrong.
117.
by jra_samba 6y ago
For EINTR it never returns a short read, as the only way to see EINTR is a return of -1 with errno==EINTR. We handle signals fine.
118.
by jra_samba 6y ago
Obviously the Windows SMB server never returns short reads, otherwise this bug would never have made it out of Redmond or Cupertino. On Linux, pread also never returns short reads against disk files if the bytes are available, which is why
119.
by jra_samba 6y ago
It was really interesting how this was found. A user started describing file corruption when copying to/from Windows with the io_uring VFS module loaded. Tests using the Linux kernel cifsfs client and the Samba libsmbclient libraries&#
120.
by jra_samba 6y ago
Also, note the preadv2 man page which has a flags field with one flag defined as: ------------------------------- RWF_NOWAIT (since Linux 4.14) Do not wait for data which is not immediately available. If this flag is specified, the preadv2(

More ›