| bio | website | google.com |
|---|---|---|
| location | Geosynchronous orbiting HQ | |
| age | 93 | |
| visits | member for | 3 years, 11 months |
| seen | yesterday | |
| stats | profile views | 74 |
Linux geek
|
Oct 17 |
comment |
Apache running as root instead of user specified in httpd.conf What you're missing is that apache is not running at all. As mentioned by Colin, the process you're seeing is the grep command you're just executing. |
|
Oct 9 |
comment |
Samba mount isn't following a symlink Note that unix extensions and wide links are mutually incompatible. Enabling one will disable the other, so do verify your configuration. As you're using a Linux client, you almost certainly want to have wide links disabled and unix extensions enabled. |
|
Oct 7 |
comment |
Clear / Flush cached memory @TheDave: Unless you specifically want to benchmark the cold cache situation, you should never use it. |
|
May 15 |
comment |
Running out of LowMem with Ubuntu PAE Kernel and 32GB of RAM Ugh, holy zombie question! Why did this turn up on the front page? |
|
Apr 19 |
comment |
Shell command slow when using pipe, fast with intermediate file I'll second the recommendation for strace. My guess would be that gs turns off IO buffering when outputting to a pipe instead of to a file. |
|
Mar 18 |
comment |
Could high MaxClient value cause CPU load issues no matter actual server stress caused by visits? @Saix: MaxRequestsPerChild has nothing to do with the number of concurrent requests the server is capable of handling. It just means that after a certain amount of requests, the child exists, thus it's useful as a workaround for memory leaks, application hangs and other bugs. You need to determine what is a suitable value for your environment. |
|
Feb 23 |
comment |
Maximum number of users logged simultaneously into Linux Note that MaxSessions, per the manpage, is "maximum number of open sessions permitted per network connection". AFAIK there is no builtin hard limit on the number of total sessions other than available resources (free ports, memory, whatever). |
|
Feb 23 |
comment |
Maximum number of users logged simultaneously into Linux @OsamaHussain: /etc/ssh/ssh_config is the config file for the ssh client. If you install the "openssh-server" package you get the server, including the /etc/ssh/sshd_config file. |
|
Dec 20 |
comment |
Linux Stealth Process Do crackware authors think that naming an executable "stealth" will make it invisible to admins? :) |
|
Nov 4 |
comment |
High system cpu load (%sys), system locks Ah, you can check e.g. /proc/softirq; unfortunately I don't know of any tool that displays individual softirqs over time. Alternatively, run something like "dstat -ar --socket --tcp" when you get a spike and post the results. |
|
Nov 4 |
comment |
High system cpu load (%sys), system locks That %si corresponds to time spent in softirq handlers? Well, if you don't believe me or some other documentation you might find, you can read the kernel source code. |
|
Oct 14 |
comment |
What exactly is a “load” when referring to node status of a Compute Cluster? @JacobHayden: Sorry, to clarify, the number of running and runnable THREADS, not processes. I don't know what you mean by the size of the processes; if you mean memory usage, then no, it doesn't affect the load average. |
|
Oct 14 |
comment |
What exactly is a “load” when referring to node status of a Compute Cluster? @JacobHayden: The number of running and runnable processes averaged over some period of time, typically, 1, 5, 15 minutes. |
|
Aug 31 |
comment |
NFS v3 versus v4 Actually, Linux also defaults to mounting with tcp |
|
Jun 13 |
comment |
Best practices for mounting Sun SAN NFS on Linux (Debian Squeeze) FWIW, Linux no longer supports the intr option. From nfs(5): "The intr / nointr mount option is deprecated after kernel 2.6.25. Only SIGKILL can interrupt a pending NFS operation on these kernels and if specified, this mount option is ignored to provide backwards compatibility with older kernels." |
|
Jun 1 |
comment |
Free alternative to NI Real-Time Hypervisor? From the words "real-time" in the name of the product, presumably it offers deterministic interrupt latency which is not something a general purpose OS offers. |
|
Apr 28 |
comment |
How to increase filename size limit on EXT3 (on ubuntu) On Linux PATH_MAX is 4096. It might be possible to get around that for some syscalls by changing cwd and using relative paths, though. That being said, a redesign to not need such long file names in the first place is perhaps the least painful approach. |
|
Apr 14 |
comment |
Task spooler for computing server on Debian slurm is another option which is quite popular, although perhaps more targeting efficient scheduling of large MPI jobs on large clusters. |
|
Dec 30 |
comment |
Cannot increase MTU on gigabit network @Sirex: 20% under what circumstances? @work I just measured 940 Mb/s using iperf without jumbo frames, which is 94% of the theoretical maximum, so a 20% improvement is just impossible in that case. (CPU usage was 5% of one core on the sender and 1% on the receiver; not only CPU's but also NIC's and the network stack in OS'es are different from 10 years ago.) |
|
Dec 29 |
comment |
Cannot increase MTU on gigabit network Or just forget about jumbo frames; it's not like they make much difference with current CPU's @1Gb. |