- luv-main - luv.asn.au

Re: IDE & Tasks Lists ~ Cross Platform
by Russell Coker 27 May '17

27 May '17

On Thursday, 25 May 2017 2:59:47 PM AEST Erik Christiansen via luv-main wrote: > > As long as they're committing their changes regularly to the version > > control system, > > I had the developers working on individual branches, then performed all > main branch commits myself, after checking that there were no CRLFs or > other nonsense, and that it compiled. Adding a useful version tag and > management-relevant commit message is also best done with an overview. One large project that I once worked on had an automated build checking system that would email the developer and their team leader about any potential build issues with code they checked in. This happened regularly due to conflicting changes and some limitations of the language and development environment (in house language that had some deficiencies in some regards but I can't remember clearly). The standard practice was to work on such issues as soon as you arrived at work and expect the team leader to ask for a progress report just before lunch. It wasn't ideal in some ways, but it worked reasonably well. > > built-in CI tools (e.g. you can configure it to automatically try to > > compile the software on every commit and report the outcome to the > > developer), and more. > > Maybe things are different in the embedded world, but I can't remember a > developer submitting code which didn't build - that would be a > professional embarrassment never lived down. And the makefile could > readily support a "make commit", to automate a pre-commit build. I have > to admit to relying on healthy paranoia to ensure I checked that it > built, before the commit. That depends on the scope of the project. If you have a project where a full build takes a few hours then developers tend to just compile the bits that they are working on which has some potential for conflicting commit. If you have a project with dependencies on external libraries then if the developers don't all have the same versions of those libraries you can have commits that don't compile for others. -- My Main Blog http://etbe.coker.com.au/ My Documents Blog http://doc.coker.com.au/

1 0

Re: IDE & Tasks Lists ~ Cross Platform
by Craig Sanders 26 May '17

26 May '17

On Thu, May 25, 2017 at 02:59:47PM +1000, luv-main(a)luv.asn.au wrote: > > and following the required coding style, > > As for "required", I let the team drive the coding style requirement, as > my only needs were consistency and readability. Since the team had set > the style standard (mostly pilfered clauses), it was self-enforcing. Any > cowboy was soon lassoed by annoyed colleagues. by style requirements, i was mostly referring to formatting like spaces, tabs, end-of-line markers etc that you mentioned. also things like where braces belong - stupidly wasting a whole screen line on a brace or putting it on the end of the line starting the block (e.g. subroutine definition) where it belongs :) and sometimes other stuff like self-documenting code with common templates for functions etc (input, output, notes including algorithm summary, known bugs and limitations, etc) > > built-in CI tools (e.g. you can configure it to automatically try to > > compile the software on every commit and report the outcome to the > > developer), and more. > > Maybe things are different in the embedded world, but I can't remember > a developer submitting code which didn't build - that would be a > professional embarrassment never lived down. And the makefile could > readily support a "make commit", to automate a pre-commit build. I > have to admit to relying on healthy paranoia to ensure I checked that > it built, before the commit. i've seen lots of developers write and submit code that compiles perfectly on their machine in their heavily customised (idiosyncratic mess) environment that fails to build anywhere else. automated tools to build and run a test suite on every checkin is amazingly useful for catching such problems early. craig -- craig sanders <cas(a)taz.net.au>

1 0

IDE & Tasks Lists ~ Cross Platform
by Piers Rowan 25 May '17

25 May '17

Hi all, We run CentOS on our servers and our dev machines are Linux or Windows (and probably a Mac somewhere but we don't like to talk about that guy!) We have grown quite a bit and having each dev running their pet dev environment seems eclectic and difficult to manage (aka manage down when you need to help a colleague and it take you 5 minutes to work out how their IDE / screen is working). I use The Eclipse for PHP IDE and one of the functions I like about it is /** @todo ... **/ creates a task and this task can be managed on some central service providers. I am seeking advice about cross platform IDE's that work well with shared tasks. Ideally something that can import messages from GMail as tasks would be great (I see another plugin that does this in Eclipse). It doesn't have to be Eclipse but the main aim is to manage individual coding tasks better. If we can get some metrics from a management point of view that would be great. If a platform can support multiple IDE's efficiently then great. Currently we use FreshDesk as out KB platform. Any ideas appreciated (as are rants/feedback) Have a great weekend. Thanks Piers

2 1

LUV server
by Russell Coker 24 May '17

24 May '17

Earlier today I changed the IPv6 address used for the LUV server. I didn't keep the old address working because almost no-one uses IPv6 and IPv6 clients can generally fall back to IPv4 if necessary, so the time that the old address remains in DNS caches shouldn't be a problem. Hetzner had decided to remove the address range they had assigned to that server and I had to apply for a new /64. On Friday afternoon the hardware hosting the LUV VM crashed for unknown reasons. A couple of hours later it booted again of it's own accord and there was nothing in the logs. So that I can fix such things faster in future I have setup 2 monitoring systems with Jabber that monitor each other. Now I just need to monitor the Jabber client on my phone (see my previous message). Some time recently an update to Dovecot had caused it to stop working correctly. I didn't notice as I don't have an IMAP account on the LUV server. To prevent that sort of thing happening again I have written a monitoring script for IMAP that will alert if there is no new mail for more than 10 minutes in a test account. I believe that some mail to the LUV president was delayed because of this. But it won't happen again. -- My Main Blog http://etbe.coker.com.au/ My Documents Blog http://doc.coker.com.au/

3 2

systemd-logind memory limit
by Russell Coker 22 May '17

22 May '17

The traditional way of limiting memory use by users is to set a per-process limit of memory use "as" in /etc/security/limits.conf and also limit the number of processes. Therefore the amount of memory that may be used is the address space limit multiplied by the number of processes. This isn't that good. For example if a user is compiling C++ software the g++ program can take quite a lot of RAM and there will also often be plenty of shell scripts etc running. The RAM requirements for a g++ process multiplied by the number of processes reasonably needed for a couple of login sessions and all the scripts for building may be more than you want to allocate to them. As an aside I'm thinking of how I killed some Unix servers while doing assignments for Unix Systems programming when I was in university. :-# https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=860410 An additional problem is that there's a bug in recent versions of sshd in that limits.conf applies to the sshd process. https://manpages.debian.org/testing/systemd/logind.conf.5.en.html When you use systemd the systemd-logind creates a new cgroup named user-$UID.slice. # cat \ /sys/fs/cgroup/memory/user.slice/user-506.slice/memory.max_usage_in_bytes 99999744 http://tinyurl.com/mhjb8ct I've set the max_usage_in_bytes to 100M (see the above Red Hat URL for an explanation of this). But it doesn't seem to work, I've written a test program that allocates memory and sets it via memset() and it gets to the ulimit setting without being stopped by the cgroup limit. Any suggestion on how to get this going properly? The next problem of course will be having systemd-logind set the limit when it creates the cgroup. Any suggestions on that will be appreciated. -- My Main Blog http://etbe.coker.com.au/ My Documents Blog http://doc.coker.com.au/

2 1

Resizing jpegs, limiting file size
by Mark Trickett 22 May '17

22 May '17

Hello All, I am wanting to cut the file size of photos from my phone. I have tried opening in GIMP, but takes a bit of mousing and clicking around, and even saving/exporting several times to get the size down. I think the imagemagik suite should be able to do, but my reading of the man pages does not make it apparent to me. They talk of resizing, but it looks like the linear extent, rather than loosing some detail of the same extent of image. I would appreciate any contributions. regards, Mark Trickett

9 10

Resizing jpegs, limiting file size
by Andrew Greig 20 May '17

20 May '17

Hi Mark, Early in the thread I suggested using Darktable to resize your jpgs from your Nokia. And at that stage your reply was that you thought it would not resize them as your input files were also jpg. Well I found the time to try it out tonight when I had to resize a rather large proof for uploading, and it does a fantastic job of resizing a jpg. And with its ability to handle batch processing you could knock the whole lot over quite quickly. I found that the the "quality" setting for the jpg was the greatest controller of file size and in the end I settled on 2000 x 1500 px at 95% quality for 660 Kb wheras at 90% the size was a bit over 200Kb, there is also a switch to prevent up-scaling. So, if you loaded all your shots into Darktable, nominated your folder for export, set the size and quality, hit Ctrl+ A, and then the export button, in no time flat (depending on RAM and CPU) you will have your result. Cheers Andrew Greig

1 0

Fwd: possible grant request for training server
by Russell Coker 15 May '17

15 May '17

I want to run more hands-on training sessions with virtual machines and I want to do some of them remotely to train people in other parts of Australia. The servers we currently have are inadequate because we have nowhere to run them 24*7. So here is the message I sent to the Linux Australia list. ---------- Forwarded Message ---------- Subject: possible grant request for training server Date: Sun, 14 May 2017, 02:39:59 PM From: Russell Coker <russell(a)coker.com.au> To: "linux-aus(a)linux.org.au" <linux-aus(a)linux.org.au> NB This is not an actual grant request. This is a discussion prior to determining what grant to request. It could be that LA members offer what is needed without the need for an LA grant request. http://tinyurl.com/pkqpgyx The German hosting company Hetzner offers a range of very affordable servers at low prices, the above URL has links for some of their best value for money offerings which includes E39 ($57.70 according to xe.com) per month for a system with 64G of RAM and 2TB of RAID-1 storage (250G of SSD isn't enough and 500G of SSD makes it too expensive). For some years my blog has been hosted on a VM on such a Hetzner server and the LUV server is on another VM on the same system. So Hetzner has been working well. In the past LUV has run a number of training sessions about system administration. This has included setting up DNS, mail servers, Cassandra, BTRFS, and ZFS. Those sessions have been well received. I am sure that other LUGs in Australia could run similar hands on training sessions. I don't think that the grant required would be justified for just LUV use, but I anticipate that other LUGs in Australia will make use of it - if only to go through the course work that I and others have prepared for LUV. But I really hope that other LUGs will develop course work too. If we had a spare server for running VMs, I would be happy to run the Dom0 and set things up to allow people from various LUGs to grant access to VMs to their members for training sessions at meetings and also for online training over IRC or other IM systems. Another possibility instead of Hetzner is that I have an offer of free hosting in Melbourne with a reasonable number of IPv4 addresses (maybe dozens of addresses as opposed to Hetzner offering few addresses and making you pay extra for them). But that free hosting requires a system that's not too noisey (I was told to remove a previous system LUV had for this purpose because of the noise). If someone has a spare system with 32G+ of RAM and isn't noisey (IE not rack mount) then that would be an option. http://www.dell.com/au/business/p/poweredge-t30/pd A Dell PowerEdge T30 can take up to 64G of RAM and starts at $1,087 (maybe less with RACV discount). I'm not sure how much it costs to expand it to 64G, the Dell web site isn't helpful in this regard. I am not set on Dell, it's just a really easy way of getting a price to start the discussion. As an aside buying a Dell will be cheaper than renting at Hetzner for 2 years. If an LA member has a tower server that's they don't need which can take a suitable amount of RAM then that would be great, CPU power is not a requirement, just RAM and storage. If such a server isn't in Victoria then I'll apply for an LA grant for courier fees. Please offer any suggestions about cheaper ways of achieving these goals. If you want to start a more general discussion about servers for running Linux then please change the subject. -- My Main Blog http://etbe.coker.com.au/ My Documents Blog http://doc.coker.com.au/ ----------------------------------------- -- My Main Blog http://etbe.coker.com.au/ My Documents Blog http://doc.coker.com.au/

1 0

Dell Server, ESX 6.5 and iSCSI
by Peter Ross 15 May '17

15 May '17

Hi all, we have new Dell Servers with Dell's VMware ESX 6.5 installed. They have multiple 10GB interfaces. With NFS we get expected speed, basically to flood the disks up to their limits. However, with iSCSI the network becomes the bottleneck, we only reach 20% of the possible performance. We installed Ubuntu 16.04 on one of the servers, for testing. With Ubuntu we are able to run iSCSI properly. We tried iSCSI with ESX against a Powerfault MD3860i and against a Synology NAS, both setups show the same slowness. So there must be something wrong with the ESX setup in this environment, I believe. So far we looked at "Best practise guides" for VMWare and tried to play with Jumbo frames, made sure they are enabled (size 9000 as recommended in these guides) on all components and disabled flowcontrol. It did not help. My colleague exhausted all options he can think of, I cannot think of anything either, and we asked VMware but did not have a good answer yet.. Of course it is urgent;-) So I wonder whether I can pick your brains. Do you have any ideas what could cause the issue? Thank you for all ideas Peter

2 2

Installing a non-free deb package into Ubuntu 16.04
by Andrew Greig 10 May '17

10 May '17

Hi, I have been using Corel Aftershot Pro 3 for Linux under OpenSuse and it is fantastic. Recently I bought a new laptop with 12Gb RAM so that I could run post production work away from my desktop, and so I have downloaded the deb package for it, but unlike my experience of RPM, the deb/Ubuntu experience is abysmal. I have tried double clicking on the package in my Downloads folder, but nothing happens. On a Ubuntu forum it was suggested that an upgrade would fix the acknowledged issue, but it hasn't. 17 years of RPM has not equipped me for this. So I would appreciate hearing where I have gone wrong and how I could get it right. Gratefully Andrew Greig

5 7