366 reputation
313
bio website
location
age
visits member for 2 years, 9 months
seen Jul 20 at 15:53

Jan
15
comment TCP: Maximum number of connections (client and server)
Ok, I've run some tests, and it seems that Linux doesn't allow source port reuse by default. When setting SO_REUSEADDR though, it is possible for the port to be reused! Of course, even then the connect() fails when connecting to the same host.
Jan
15
comment Why use wifi channels other than 1, 6 or 11?
@DavidSchwartz - important point. The reference by Fiasco Labs seems to contain the results by Cisco you mention. See also superuser.com/questions/443178/….
Nov
22
comment TCP: Maximum number of connections (client and server)
I already suspected that the practical limits are not imposed by the specs, but by the implementations - but I would like to learn those as well :)
Nov
1
comment Modern GPUs: How “intelligent” are they?
Wow, amazing answer, thanks!
Sep
25
comment Why does this line appear every time I launch iTerm on my MacBook?
The string might not be included in the script that way - it's quite possible that it's created dynamically from your home directory (something like echo . $HOME/Dropbox/private/autosync").
Sep
25
comment How long can file system writes be cached with ext4?
I'm pretty sure he meant the former, the latter would make ext4 on Linux a pretty dangerous file system to use ;) The script looks like a nice workaround; I'll give it a try and maybe evaluate some of my most important applications with strace - maybe they are all using fsync(), and I'm worrying too much...
Sep
25
comment How long can file system writes be cached with ext4?
Really? In bugs.launchpad.net/ubuntu/+source/linux/+bug/317781/comments/45 it is specifically mentioned that unallocated pages will NOT be written to disk on a commit (but of course on an fsync()). The patch fixes some common cases where that behavior is problematic by forcing allocation; however, there is nothing said about overwriting data.
Sep
25
comment How long can file system writes be cached with ext4?
Thanks, the "ALL dirty data" part was exactly what I was worrying about! I was worried that there were more exceptions in addition to delayed allocation (which can cause new data to remain in the write cache even after the commit interval).
Sep
25
comment How long can file system writes be cached with ext4?
I know about fsync(); I'm asking as a user of applications that might or might not use it. I've updated my question.
Sep
15
comment (How) does deleting open files on Linux and a FAT file system work?
@Scott, yes, that is exactly what I meant. It seems that Linux actually does that, because when I uncleanly unmount a FAT file system with opened but deleted files on it, there is always some corruption detected by fsck.vfat, and it always is about blocks not marked as free but also not part of any file - which would suggest that the directory entry is deleted, but the corresponding blocks in the FAT are not set to show up as free space until the last handle is closed.
May
10
comment Does 802.1X and RADIUS prevent rogue APs?
The AP is not the authentication server, and only the authentication server knows the certificate.
Apr
28
comment Shell SIGKILL Keybinding
I guess he's not actually talking about the line editor - if he's typing into the shell, there would be no running foreground process, right? The shell would never actually see the keypresses if there is a foreground process running and connected to the current tty.
Apr
5
comment TCP and fair bandwidth sharing
But isn't that how TCP itself handles congestion control? Packets are dropped, and the sender throttles back. I just want the packets to be dropped "fairly", i.e. if one user consumes more than his fair share, only (or primarily) his packets will be dropped, and only his flows will have to throttle back. Of course that would imply metering each user's current packet rate, and dropping accordingly. Is that possible?
Apr
4
comment TCP and fair bandwidth sharing
But I can always drop that packet, thereby forcing TCP to throttle back, right? I'm hoping that the offending flows are TCP or TCP-like regarding their congestion control.
Jan
31
comment Why does NAT64 require DNS rewriting?
So that scenario was not considered when IPv6 was designed (or considered not to be relevant to the core standard), and DNS64 is the hack that makes it work? Or am I missing something? Would defaulting to that prefix in case no IPv4 interface is available by all IPv6 hosts break anything?
Jan
31
comment Why does NAT64 require DNS rewriting?
But that's 6to4, right? There would be no way for the hosts in the network to have "native", routable IPv6 addresses.
Jan
30
comment Why does NAT64 require DNS rewriting?
Ok, so I guess there is a pretty good reason why it is being used, but just out of curiosity: Why is there no well-known prefix? Hosts could automatically select it in case there is no IPv4 route to the target host, and all the DNS hassle could be avoided. (Hardwired IP addresses would still work, too.) Or would that interfere with the sockets API (after all, a requested IPv4 connection would suddenly be an IPv6 one)?
Nov
7
comment How does a modern browser handle scrolling?
Does that mean that the API user only has to worry about redrawing the parts that have "scrolled into view", and the OS moves all the remaining parts up/down without any application layer repaints?