226 reputation
18
bio website
location
age
visits member for 3 years, 6 months
seen Nov 13 at 12:54

Jul
11
awarded  Curious
Jun
16
asked Simultaneous iptables POSTROUTING for SNAT and MASQUERADE block outgoing ssh
Mar
10
awarded  Popular Question
Nov
20
comment Sync LVM snapshots to backup server
The link to the Zumastor project is outdated, I guess this is the correct one: shapor.com/zumastor.org
Oct
10
comment How to configure Single-Server Private Network in Xenserver without using VLAN?
Thank you, that is the way to create it in Xencenter. But it does not provide any configuration options. The point of my question is, if there is a Xenserver/Xencenter way of configuring (eg. set an IP for the Xen host) the Single-Server network. I am currently only able to do this, if I set up the network as an external network/VLAN instead of the Single-Server option.
Sep
30
asked How to configure Single-Server Private Network in Xenserver without using VLAN?
Sep
16
accepted How to keep iptables organized in Xenserver?
Sep
15
asked How to keep iptables organized in Xenserver?
Jun
17
comment How to free PHP-FPM memory?
It is true that the memory consumption of long-running php-fpm processes increases. But the ratio you are describing sounds far too much. I would suggest checking your scripts and included php modules for memory leaks. Potential culprits might include low-level stuff like xdebug or some unfortunate situations with code caches like APC.
Jan
28
comment Is there an IP v4 address that can never be assigned?
@Aaron Well, I am asking if it exists because it would make my life easier, and most importantly being consistent across all clients everywhere. In Linux after all, it is normal to pipe things into /dev/null or to have iptables reject almost everything. So the intuition for my question was, whether something like /dev/null exists for IP. Unfortunately even the RFC you posted regarding routing the space reserved for documentation is not generally followed.
Jan
23
comment Is there an IP v4 address that can never be assigned?
300.300.300.300 may or may not work. It is an invalid address, which is nice because it can never be reached and all requests fail immediately, but it will fail DNS validation and may or may not produce unforeseen results.
Jan
22
comment Is there an IP v4 address that can never be assigned?
@Aaron This is one of the tasks I am trying to accomplish and unfortunately I have reasons not to use localhost. It is also about temporarily pointing scripts to surely non-existing addresses, in that case having a non-existing but valid address allows me to make sure that services do not work, while I can simply use copy and paste instead of relying on a service staying turned-off. The 0.0.0.0 suggestion by HopelessN00b and you makes sense by reading the docs, but in practice it resolves to localhost for me (using debian and tinydns, the DNS reply for type A is correct 0.0.0.0).
Jan
21
awarded  Yearling
Jan
21
accepted Is there an IP v4 address that can never be assigned?
Jan
21
comment Is there an IP v4 address that can never be assigned?
My question is, if there are address spaces that are not routed by routers, are there also address spaces that are generally not answered hosts? My usage case is simple: Assume a domain or hostname which you would like to temporarily route to "nowhere" without leaving A records in DNS blank. And you would like to make sure it does not resolve to some local network of an unaware visitor.
Jan
21
asked Is there an IP v4 address that can never be assigned?
Dec
10
awarded  Tumbleweed
Nov
21
comment Keeping configs, user ids, groups and passwords consistent across Linux-servers
@EEAA Yes, the version control is probably the best thing about it. I have decdided to try cdist, it seems to be the most lightweight software i found.
Nov
21
accepted Keeping configs, user ids, groups and passwords consistent across Linux-servers
Nov
20
comment Keeping configs, user ids, groups and passwords consistent across Linux-servers
Thank you for this answer. I am looking at the configuration systems currently, and I agree that your answer is probably that best solution for my question. Still it seems to me, that I am just pushing the problem to another layer. To use a config system, I have to install it and a user for it, make sure that it does not mess interfere with existing numerical ids I use across servers, store an shh key or other means of authentication... But basically this is not that much more than I needed before.