Reputation
868
Next privilege 1,000 Rep.
See votes, expandable usercard
Badges
10 23
Newest
 Caucus
Impact
~278k people reached

May
1
comment How can I make an internet facing TFTP server secure?
From wikipedia: "Due to the lack of security, it is dangerous to use it over the Internet. Thus, TFTP is generally only used on private, local networks."
Apr
28
comment How to investigate the cause of a 100% CPU event that lasted for hours?
It's unlikely to happen again, at least in a reasonable time frame that would allow some closure. It didn't happen for about a year of that server running that configuration, and so far hasn't happened since. I was hoping to get some advice on to what extent it's possible to investigate something that happened in the past. Have I done all that's possible?
Apr
28
comment How to investigate the cause of a 100% CPU event that lasted for hours?
I've added some more info to the question.
Apr
28
comment How to investigate the cause of a 100% CPU event that lasted for hours?
Disk and memory usage were normal/unaffected during that time as shown in other charts (not pictured). The charts are compiled by a script that runs every 5 minutes installed on the VPS by default by the VPS provider. They report CPU, memory, disk and network usage every 5 minutes. The CPU is the only one that shows an anomaly. I have a daily backup that runs a couple of hours prior to the start of this "event". It seemed to run normally although it did cause a higher server load (but not CPU utilisation) than normal, pointing to a possible I/O congestion issue.
Jan
24
comment Why nginx gzip_http_version defaults to HTTP 1.1
The first sentence has been removed from newer versions of the documentation, so presumably it does now still send the Vary header. But the impossibility of using keep-alive when gzipping on HTTP 1.0 (because the content-length cannot be known prior to output) still exists and there's no way around that. So decide which is worse: HTTP 1.0 clients missing out on gzip, or HTTP 1.0 clients missing out on keep-alive (which often isn't possible for dynamic scripts on HTTP 1.0 anyway).
Jan
24
comment Why nginx gzip_http_version defaults to HTTP 1.1
This setting does not control what HTTP version is used, it controls whether gzip should be disabled for HTTP 1.0 requests.
Jun
28
comment dspam vs spamassassin vs bogofilter
I have since found a solution in proxsmtp.
Jun
20
comment Do you use postfix or qmail and why
I thought that the "weird licensing" was rectified in 2007 when it was donated to public domain. I think the lack of updates is more of a problem than the licensing these days in terms of acceptance by distros.
Apr
22
comment BIND - why duplicate nameserver entries (@ and *)?
Well, assuming this is mainly about wanting to use subdomains per service, like "www" for websites, that's an opinion that the rest of the world is increasingly moving away from, including for example this website (serverfault.com). Relevant discussion also here and here.
Apr
22
comment BIND - why duplicate nameserver entries (@ and *)?
Redirects? You mean a CNAME? That's not possible for a domain that has any other records and not a good idea otherwise. Or do you mean an HTTP redirect? Those are nothing to do with DNS, bind, or zone files.
Apr
5
comment Is NTFS 16 TB volume size limit “real”?
@Massimo that's what people said about 20GB drives once. And they would have said it about 500MB drives once, too. But I am from the future, and I have two 3TB drives, and yes it's impractical to copy 3TB of data but how else are you gonna store 3TB of data?
Apr
5
comment Is NTFS 16 TB volume size limit “real”?
@jrtipton NTFS itself is designed to support volumes as big as 2^64 - 1 clusters of up to 64kB each. That is, 1,099,511,627,776 TB (over one trillion Terabytes!). However, the NTFS support in Windows (since Windows XP) only supports 2^32 - 1 clusters of up to 64kB each. That is, 256 TB.
Feb
20
comment Updating production Ubuntu boxes the dos and don'ts
The debian/ubuntu utility checkrestart is very useful in determining which processes have been updated but still need to be stopped and restarted in order to get the new code.
Feb
15
comment Storing a million images in the filesystem
Because it's not a universally bad solution. Sure, for most people I wouldn't recommend it, but there is a small subset of applications for which the benefits (yes there are some benefits) are actually quite important, like acid compliance, or replication, or as someone else mentioned having versioning. Though admittedly, this particular answer really didn't sell it very well and seemed to be recommending it for a the wrong reasons.
Jan
31
comment Can Ubuntu Desktop “become” Ubuntu Server?
No even at 10.04 this wasn't correct - it has always been easy to change kernels.
Dec
13
comment BIND - why duplicate nameserver entries (@ and *)?
This doesn't really answer the question though...
Dec
10
comment Can Ubuntu Desktop “become” Ubuntu Server?
Note to anyone visiting this page: this answer is wrong. Firstly, it's easy to change kernels. Secondly, there are no longer separate kernels for server: linux-image-server is now just an alias for the generic kernel.
Dec
7
comment How to minimise SpamAssassin (spamd) memory use
From my observation and using the tool smem, it looks like spamassassin uses around 50 MB of RAM, and that if you fork it into multiple processes, almost all their memory is shared memory, so it'll still use around 50 MB of RAM total amongst all processes, even though ps reports each one having a RSS of 50 MB. YMMV.
Dec
4
comment Recommended DNS SOA record TTL default?
To clarify, the SOA Minimum TTL field stores the TTL value to be used to cache a negative request - a request made to the zone for some resource which doesn't exist. Their explanation is sort of true but fails to clarify it's only for negative responses. Secondly, the SOA Refresh is never used by normal DNS queries, it's only used in situations where you have secondary (slave) nameservers updating themselves from your primary (master) nameserver. So their explanation of that field is definitely untrue.
Nov
30
comment Why should one have a secondary DNS server?
I've read quite a few comments that indicate that many real-world DNS caching services (like those used by ISP) don't re-try using a second name server, they just fail if the first server doesn't respond. For example, this answer on serverfault. In which case, if you have two separate nameservers you need to make sure both are up, because either one being down can lead to downtime for the hosted domains. This does go against common practice and the RFCs, but seems concerning.