From zanchey at ucc.gu.uwa.edu.au Tue Mar 3 15:14:09 2015 From: zanchey at ucc.gu.uwa.edu.au (David Adam) Date: Tue, 3 Mar 2015 15:14:09 +0800 (AWST) Subject: [tech] the Xserve Message-ID: [BG3] arranged for the donation of an old Intel Xserve (1,1) which we have been trying to get running. It boots and the external VGA adapter works fine, so now all we need to do is get an OS onto it. Unfortunately it refuses to boot the iMac 10.6 discs (apparently requiring 10.6 Server Edition) and AFAICT it's not possible to buy any 10.7 media any longer. apparently there is no separate Xserve media for 10.7. If anyone has either 10.6 Xserve or 10.7 installation discs that they could provide to UCC we'd like to get an OS X server up and going. David Adam zanchey at ucc.gu.uwa.edu.au From mjpomery at ucc.asn.au Wed Mar 4 00:00:05 2015 From: mjpomery at ucc.asn.au (Mitchell Pomery) Date: Wed, 4 Mar 2015 00:00:05 +0800 (AWST) Subject: [tech] Vending Machine 3 Button & MIFARE Reader Message-ID: The Three Button: As we all know, the vending machine's 3 button is the worst thing that can ever happen to you. But worry no more, for I had an ingenious idea that worked! What was the problem: The three button the the vending machine bounced, meaning it was not uncommon to get four 3's when you only wanted one. This was the state it got two when I finally got fed up with it. How I tried to fix it: I tried to impliment hardware debouncing, but found it largely unsuccessful due to the type of button and the inability to rewire it nicely. What I ended up doing: Desoldering both the 3 and the reset keys, swapping the switches then soldering them back in. What the end result was: I was expecting the reset key to bounce, but it did not. The three button did not bounce either. I somehow managed to take a bouncy switch and a non bouncy switch and create two non bouncy switches. MIFARE Reader: The new MIFARE reader is capable of reading cards much faster than the previous one was. This meant that people enrolling new cards would see "Card Already Enrolled" when they first held it up. This did confuse some of the new members. Since then I have patched VendServer to only read each card once. When you hold a card up to the reader, it will be read multiple times, but the software will only act upon that card once. Using another card or logging out of the machine is enough to get it read again. And while typing this, I have thought of a potential bug: Holding a card up that is not linked to an account. I will investigate this sometime tomorrow. The Eternally Busy, Mitchell Pomery [BG3] From mjpomery at ucc.asn.au Wed Mar 4 14:53:21 2015 From: mjpomery at ucc.asn.au (mjpomery at ucc.asn.au) Date: Wed, 04 Mar 2015 14:53:21 +0800 Subject: [tech] Vending Machine 3 Button & MIFARE Reader In-Reply-To: References: Message-ID: Hi Mark, As per my email last night, attached is a patch to fix the bug introduced with the fix to MIFARE card reading in the Vending Machine. The card reader won't read a card that isn't linked to an account when you log into the machine to enrol it, if it was the last card tried. Hi Door, The Important Bit (For Door members who skim this): It also seems that there is an issue with reading some student cards. They are not picked up by the card reader at all (so it isn't even getting to my code changes). I haven't seen it personally, so if anyone sees this happening, can they make a note of who it was if I'm not around the clubroom, or grab me if I am there. Regards, Mitch Pomery UCC President 2014 On 2015-03-04 00:00, Mitchell Pomery wrote: > The Three Button: > > As we all know, the vending machine's 3 button is the worst thing that > can > ever happen to you. But worry no more, for I had an ingenious idea that > worked! > > What was the problem: > The three button the the vending machine bounced, meaning it was not > uncommon to get four 3's when you only wanted one. This was the state > it > got two when I finally got fed up with it. > > How I tried to fix it: > I tried to impliment hardware debouncing, but found it largely > unsuccessful due to the type of button and the inability to rewire it > nicely. > > What I ended up doing: > Desoldering both the 3 and the reset keys, swapping the switches then > soldering them back in. > > What the end result was: > I was expecting the reset key to bounce, but it did not. The three > button > did not bounce either. I somehow managed to take a bouncy switch and a > non > bouncy switch and create two non bouncy switches. > > MIFARE Reader: > > The new MIFARE reader is capable of reading cards much faster than the > previous one was. This meant that people enrolling new cards would see > "Card Already Enrolled" when they first held it up. This did confuse > some > of the new members. Since then I have patched VendServer to only read > each > card once. When you hold a card up to the reader, it will be read > multiple > times, but the software will only act upon that card once. Using > another > card or logging out of the machine is enough to get it read again. > > And while typing this, I have thought of a potential bug: Holding a > card > up that is not linked to an account. I will investigate this sometime > tomorrow. > > The Eternally Busy, > Mitchell Pomery [BG3] > _______________________________________________ > List Archives: http://lists.ucc.gu.uwa.edu.au/pipermail/tech > > Unsubscribe here: > http://lists.ucc.gu.uwa.edu.au/mailman/options/tech/bobgeorge33%40ucc.asn.au -------------- next part -------------- A non-text attachment was scrubbed... Name: MIFARE_bug.patch Type: text/x-diff Size: 1514 bytes Desc: not available Url : http://lists.ucc.gu.uwa.edu.au/pipermail/tech/attachments/20150304/1859171e/attachment-0001.patch From trs80 at ucc.gu.uwa.edu.au Wed Mar 4 22:12:44 2015 From: trs80 at ucc.gu.uwa.edu.au (James Andrewartha) Date: Wed, 4 Mar 2015 22:12:44 +0800 (AWST) Subject: [tech] the Xserve In-Reply-To: References: Message-ID: On Tue, 3 Mar 2015, David Adam wrote: > Unfortunately it refuses to boot the iMac 10.6 discs (apparently requiring > 10.6 Server Edition) and AFAICT it's not possible to buy any 10.7 media > any longer. apparently there is no separate Xserve media for 10.7. > > If anyone has either 10.6 Xserve or 10.7 installation discs that they > could provide to UCC we'd like to get an OS X server up and going. I asked the Apple guy at work, and he said there's no such thing as a 10.7 install disc, but he did lend me a usb spaceman with the installer. I tried booting, but it got video corruption towards the end. -- # TRS-80 trs80(a)ucc.gu.uwa.edu.au #/ "Otherwise Bub here will do \ # UCC Wheel Member http://trs80.ucc.asn.au/ #| what squirrels do best | [ "There's nobody getting rich writing ]| -- Collect and hide your | [ software that I know of" -- Bill Gates, 1980 ]\ nuts." -- Acid Reflux #231 / From zanchey at ucc.gu.uwa.edu.au Wed Mar 4 22:21:55 2015 From: zanchey at ucc.gu.uwa.edu.au (David Adam) Date: Wed, 4 Mar 2015 22:21:55 +0800 (AWST) Subject: [tech] the Xserve In-Reply-To: References: Message-ID: On Wed, 4 Mar 2015, James Andrewartha wrote: > On Tue, 3 Mar 2015, David Adam wrote: > > > Unfortunately it refuses to boot the iMac 10.6 discs (apparently requiring > > 10.6 Server Edition) and AFAICT it's not possible to buy any 10.7 media > > any longer. apparently there is no separate Xserve media for 10.7. > > > > If anyone has either 10.6 Xserve or 10.7 installation discs that they > > could provide to UCC we'd like to get an OS X server up and going. > > I asked the Apple guy at work, and he said there's no such thing as a 10.7 > install disc, but he did lend me a usb spaceman with the installer. I > tried booting, but it got video corruption towards the end. Target disc mode might be the way to go then? Unfortunately Napoli isn't booting. Again. So we will require someone with a Mac laptop and a Firewire cable - the ones UCC have are all FireWire 400 8-pin to 4-pin, which are enormously unsuitable. [DAA] From ucc at cueballcentral.com Thu Mar 5 03:01:01 2015 From: ucc at cueballcentral.com (Mark Glossop) Date: Thu, 5 Mar 2015 03:01:01 +0800 Subject: [tech] the Xserve In-Reply-To: References: Message-ID: <15DD3D12-DB74-43AB-BD0D-53B5EB63ADCB@cueballcentral.com> On 4 Mar 2015, at 22:21, David Adam wrote: > > On Wed, 4 Mar 2015, James Andrewartha wrote: >> On Tue, 3 Mar 2015, David Adam wrote: >> >>> Unfortunately it refuses to boot the iMac 10.6 discs (apparently requiring >>> 10.6 Server Edition) and AFAICT it's not possible to buy any 10.7 media >>> any longer. apparently there is no separate Xserve media for 10.7. >>> >>> If anyone has either 10.6 Xserve or 10.7 installation discs that they >>> could provide to UCC we'd like to get an OS X server up and going. >> >> I asked the Apple guy at work, and he said there's no such thing as a 10.7 >> install disc, but he did lend me a usb spaceman with the installer. I >> tried booting, but it got video corruption towards the end. > > Target disc mode might be the way to go then? > > Unfortunately Napoli isn't booting. Again. So we will require someone with > a Mac laptop and a Firewire cable - the ones UCC have are all FireWire 400 > 8-pin to 4-pin, which are enormously unsuitable. > > [DAA] Xserve1,1 came with 10.4.8 (server), it _should_ boot anything from 10.4.8 Server through 10.7.x. The original install discs of that era were probably "halo" builds of 10.4.8, so they'll only work on the Xserve (and "normal" 10.4.8 builds won't likely include the specific HW support that it needs.) That said, for 10.5 and above, IIRC it doesn't _actually_ need to be Mac OS X Server installed. (I'm guessing that it won't take anything higher maybe due to it only having a 32-bit EFI? IDK...) Poking around, I found a 10.5 install DVD, a 6.6GB DVD image (.dmg) for installing 10.6, and the 10.7 Install .app (4.1GB). They should all be hardware-independent installers. No Mac OS X Server installers though #sorrynotsorry. Let me know if any of the above are of interest. Alternately, there are folks at IS/(ITS/whatever acronym it goes by this week) who may very well have the exact discs you need - I can possibly put you in touch with them if you don't already know who they might be :-) Cheers, M. -- Mark Glossop E: ucc at cueballcentral.com W: http://cueballcentral.com T: http://twitter.com/Cueball_AU From mjpomery at ucc.asn.au Thu Mar 5 19:08:38 2015 From: mjpomery at ucc.asn.au (Mitchell Pomery) Date: Thu, 5 Mar 2015 19:08:38 +0800 (AWST) Subject: [tech] uccvend-vendserver patch - VendServer.py executable on install Message-ID: Hi Mark, I've found a way to make sure that VendServer.py is an executable on install, mark all the python files as executable. This is an interim measure until I can figure out what setuptools is doing. Speaking of setuptools, is it really required, or can we roll our own utility that is nicer and does exactly what we need? It will also reduce our dependancies. Thanks, Mitchell Pomery UCC President 2014 -------------- next part -------------- A non-text attachment was scrubbed... Name: fix_file_permissions.patch Type: text/x-diff Size: 2270 bytes Desc: Url : http://lists.ucc.gu.uwa.edu.au/pipermail/tech/attachments/20150305/0b8d6c6f/attachment.patch From smspillaz at gmail.com Thu Mar 5 20:01:09 2015 From: smspillaz at gmail.com (Sam Spilsbury) Date: Thu, 5 Mar 2015 20:01:09 +0800 Subject: [tech] uccvend-vendserver patch - VendServer.py executable on install In-Reply-To: References: Message-ID: Hi, On Thu, Mar 5, 2015 at 7:08 PM, Mitchell Pomery wrote: > Hi Mark, > > I've found a way to make sure that VendServer.py is an executable on > install, mark all the python files as executable. This is an interim measure > until I can figure out what setuptools is doing. I don't mean to be nosy (and I'm making some assumptions about the particular tool at issue here), but I just saw this one fly past my inbox and thought some feedback might be helpful. Making every python file in the distribution is probably not what you want - most python files aren't supposed to be executed directly. > > Speaking of setuptools, is it really required, or can we roll our own > utility that is nicer and does exactly what we need? It will also reduce our > dependancies. Setuptools has an entry_points option where you can dynamically create a script to launch your application[1]. For example, you could have: entry_points=[ "console_scripts": [ "vend-server=VendServer.VendingMachine:main" ] ] That would run the main() function on execution of the "vend-server" command after installation. Hope that's useful :) Sam. [1] http://setuptools.readthedocs.org/en/latest/setuptools.html#automatic-script-creation > > Thanks, > Mitchell Pomery > UCC President 2014 > _______________________________________________ > List Archives: http://lists.ucc.gu.uwa.edu.au/pipermail/tech > > Unsubscribe here: > http://lists.ucc.gu.uwa.edu.au/mailman/options/tech/smspillaz%40gmail.com -- Sam Spilsbury From mjpomery at ucc.asn.au Wed Mar 18 01:04:13 2015 From: mjpomery at ucc.asn.au (Mitchell Pomery) Date: Wed, 18 Mar 2015 01:04:13 +0800 (AWST) Subject: [tech] virtualsnack patches - Error Handling and Virtual snack machine Message-ID: Hi Mark, Attached are two patches for virtualsnack. The first improves the error handling, meaning we don't catch all errors and call them one thing. It also improves the reconnect logic for the machine. The second adds an interface to the right of the screen that spins as an item is dispensed. I will make it look fancier in an upcoming commit. Regards, Mitch -------------- next part -------------- A non-text attachment was scrubbed... Name: 0002-Add-virtual-vending-machine.patch Type: text/x-diff Size: 2562 bytes Desc: Url : http://lists.ucc.gu.uwa.edu.au/pipermail/tech/attachments/20150318/7c3c85a9/attachment-0002.patch -------------- next part -------------- A non-text attachment was scrubbed... Name: 0001-Fix-error-handling.patch Type: text/x-diff Size: 7984 bytes Desc: Url : http://lists.ucc.gu.uwa.edu.au/pipermail/tech/attachments/20150318/7c3c85a9/attachment-0003.patch From mtearle at ucc.asn.au Wed Mar 4 21:21:17 2015 From: mtearle at ucc.asn.au (Mark Tearle) Date: Wed, 04 Mar 2015 21:21:17 +0800 Subject: [tech] Vending Machine 3 Button & MIFARE Reader In-Reply-To: References: Message-ID: <1425475277.2069587.235388449.6EAC7B1E@webmail.messagingengine.com> Hi Mitch I've applied and push the patch to the git repository. I'll leave you to update the running copy. Mark -- Mark Tearle On Wed, Mar 4, 2015, at 02:53 PM, mjpomery at ucc.asn.au wrote: > Hi Mark, > > As per my email last night, attached is a patch to fix the bug > introduced with the fix to MIFARE card reading in the Vending Machine. > The card reader won't read a card that isn't linked to an account when > you log into the machine to enrol it, if it was the last card tried. > > Hi Door, > > The Important Bit (For Door members who skim this): > It also seems that there is an issue with reading some student cards. > They are not picked up by the card reader at all (so it isn't even > getting to my code changes). I haven't seen it personally, so if anyone > sees this happening, can they make a note of who it was if I'm not > around the clubroom, or grab me if I am there. > > Regards, > Mitch Pomery > UCC President 2014 > > > On 2015-03-04 00:00, Mitchell Pomery wrote: > > The Three Button: > > > > As we all know, the vending machine's 3 button is the worst thing that > > can > > ever happen to you. But worry no more, for I had an ingenious idea that > > worked! > > > > What was the problem: > > The three button the the vending machine bounced, meaning it was not > > uncommon to get four 3's when you only wanted one. This was the state > > it > > got two when I finally got fed up with it. > > > > How I tried to fix it: > > I tried to impliment hardware debouncing, but found it largely > > unsuccessful due to the type of button and the inability to rewire it > > nicely. > > > > What I ended up doing: > > Desoldering both the 3 and the reset keys, swapping the switches then > > soldering them back in. > > > > What the end result was: > > I was expecting the reset key to bounce, but it did not. The three > > button > > did not bounce either. I somehow managed to take a bouncy switch and a > > non > > bouncy switch and create two non bouncy switches. > > > > MIFARE Reader: > > > > The new MIFARE reader is capable of reading cards much faster than the > > previous one was. This meant that people enrolling new cards would see > > "Card Already Enrolled" when they first held it up. This did confuse > > some > > of the new members. Since then I have patched VendServer to only read > > each > > card once. When you hold a card up to the reader, it will be read > > multiple > > times, but the software will only act upon that card once. Using > > another > > card or logging out of the machine is enough to get it read again. > > > > And while typing this, I have thought of a potential bug: Holding a > > card > > up that is not linked to an account. I will investigate this sometime > > tomorrow. > > > > The Eternally Busy, > > Mitchell Pomery [BG3] > > _______________________________________________ > > List Archives: http://lists.ucc.gu.uwa.edu.au/pipermail/tech > > > > Unsubscribe here: > > http://lists.ucc.gu.uwa.edu.au/mailman/options/tech/bobgeorge33%40ucc.asn.au > Email had 1 attachment: > + MIFARE_bug.patch > 2k (text/x-diff) From mtearle at ucc.asn.au Sat Mar 7 21:34:54 2015 From: mtearle at ucc.asn.au (Mark Tearle) Date: Sat, 07 Mar 2015 21:34:54 +0800 Subject: [tech] uccvend-vendserver patch - VendServer.py executable on install In-Reply-To: References: Message-ID: <1425735294.3705768.237046973.2BFBA912@webmail.messagingengine.com> Hi Sam > > Speaking of setuptools, is it really required, or can we roll our own > > utility that is nicer and does exactly what we need? It will also reduce our > > dependancies. > > Setuptools has an entry_points option where you can dynamically create > a script to launch your application[1]. For example, you could have: > > entry_points=[ > "console_scripts": [ > "vend-server=VendServer.VendingMachine:main" > ] > ] > > That would run the main() function on execution of the "vend-server" > command after installation. > > Hope that's useful :) > > Sam. > > [1] > http://setuptools.readthedocs.org/en/latest/setuptools.html#automatic-script-creation > Thanks, that was just the code snippet I needed to point me in the right direction. I've applied the patch attached to this email to repository for the VendServer. Mark PS. Mitch, do you want to roll it out to production? -- Mark Tearle -------------- next part -------------- A non-text attachment was scrubbed... Name: 0001-Add-entrypoint-to-setup.py-for-vendserver.patch Type: text/x-patch Size: 1782 bytes Desc: not available Url : http://lists.ucc.gu.uwa.edu.au/pipermail/tech/attachments/20150307/aeff8dd3/attachment.bin From mtearle at ucc.asn.au Sun Mar 15 20:31:19 2015 From: mtearle at ucc.asn.au (Mark Tearle) Date: Sun, 15 Mar 2015 20:31:19 +0800 Subject: [tech] Announcing virtualcoke Message-ID: <1426422679.3820073.240575061.201924E4@webmail.messagingengine.com> Hi folks Continuing my development of tools to help us work and develop on coke/snack related things, I've written a Curses simulator of the coke machine PLC. It can be found here: https://github.com/ucc/virtualcoke Mark -- Mark Tearle From trs80 at ucc.gu.uwa.edu.au Wed Mar 18 20:52:19 2015 From: trs80 at ucc.gu.uwa.edu.au (James Andrewartha) Date: Wed, 18 Mar 2015 20:52:19 +0800 (AWST) Subject: [tech] the Xserve In-Reply-To: References: Message-ID: On Wed, 4 Mar 2015, David Adam wrote: > Unfortunately Napoli isn't booting. Again. So we will require someone with > a Mac laptop and a Firewire cable - the ones UCC have are all FireWire 400 > 8-pin to 4-pin, which are enormously unsuitable. Napoli's disk is munted, and Disk Utility is unable to repair it. I had a shot at opening it up to replace the hard drive, but didn't have time to do it completely and was nearly defeated by the display cable. I've created a new Lion installer on a 4GB NetApp branded USB drive and left it plugged into the Xserve. I couldn't get it to boot any better, only once did I get the installer, and the video was still corrupted. If you are creating a new installer, 8GB is better otherwise you'll have to use an Lion DiskMaker 1.7 and run agent from Terminal at Lion DiskMaker 1.7/Lion DiskMaker.app/Contents/MacOS as it doesn't like launching directly from Finder in Mavericks. -- # TRS-80 trs80(a)ucc.gu.uwa.edu.au #/ "Otherwise Bub here will do \ # UCC Wheel Member http://trs80.ucc.asn.au/ #| what squirrels do best | [ "There's nobody getting rich writing ]| -- Collect and hide your | [ software that I know of" -- Bill Gates, 1980 ]\ nuts." -- Acid Reflux #231 / From zanchey at ucc.gu.uwa.edu.au Thu Mar 19 13:12:52 2015 From: zanchey at ucc.gu.uwa.edu.au (David Adam) Date: Thu, 19 Mar 2015 13:12:52 +0800 Subject: [tech] the Xserve In-Reply-To: References: Message-ID: On 2015-03-04 22:21, David Adam wrote: > On Wed, 4 Mar 2015, James Andrewartha wrote: >> On Tue, 3 Mar 2015, David Adam wrote: >> >> > Unfortunately it refuses to boot the iMac 10.6 discs (apparently requiring >> > 10.6 Server Edition) and AFAICT it's not possible to buy any 10.7 media >> > any longer. apparently there is no separate Xserve media for 10.7. >> > >> > If anyone has either 10.6 Xserve or 10.7 installation discs that they >> > could provide to UCC we'd like to get an OS X server up and going. >> >> I asked the Apple guy at work, and he said there's no such thing as a >> 10.7 >> install disc, but he did lend me a usb spaceman with the installer. I >> tried booting, but it got video corruption towards the end. > > Target disc mode might be the way to go then? [RME] gave this a crack over the weekend - he installed the 10.7 installation image on an extra partition and tried booting off it. As you discovered - video corruption was a problem and there definitely seems to be something strange going on hardware-wise. If we can find a spare video daughterboard that might be the way to go, but I am once again out of energy for Mac stuff. [DAA] From zanchey at ucc.gu.uwa.edu.au Fri Mar 20 16:43:29 2015 From: zanchey at ucc.gu.uwa.edu.au (David Adam) Date: Fri, 20 Mar 2015 16:43:29 +0800 (AWST) Subject: [tech] Heathred upgraded to Debian 8 Jessie Message-ID: I want to run a VM that requires a new kernel, so I upgraded Heathred to Debian 8.0 (Jessie) testing version. Nothing seems to be broken but I haven't rebooted yet - I'll leave that until a time when I can get in and fix things if need be. [DAA] From mjpomery at ucc.asn.au Wed Mar 25 15:16:24 2015 From: mjpomery at ucc.asn.au (Mitchell Pomery) Date: Wed, 25 Mar 2015 15:16:24 +0800 (AWST) Subject: [tech] virtualsnack patches Message-ID: Attached are my latest run of patches for virtualsnack. The big change made by these commits are the introduction of a virtual vending machine (making the UI look pretty). Thanks, Mitch -------------- next part -------------- A non-text attachment was scrubbed... Name: patches.patch Type: text/x-diff Size: 17071 bytes Desc: Url : http://lists.ucc.gu.uwa.edu.au/pipermail/tech/attachments/20150325/bc8df2b3/attachment-0001.patch From mtearle at ucc.asn.au Wed Mar 25 22:21:32 2015 From: mtearle at ucc.asn.au (Mark Tearle) Date: Wed, 25 Mar 2015 22:21:32 +0800 Subject: [tech] virtualsnack patches In-Reply-To: References: Message-ID: <1427293292.334347.245070041.7D787921@webmail.messagingengine.com> Hi Mitch Applied, but still some Unicode issues to sort out. Mark -- Mark Tearle On Wed, Mar 25, 2015, at 03:16 PM, Mitchell Pomery wrote: > Attached are my latest run of patches for virtualsnack. > > The big change made by these commits are the introduction of a virtual > vending machine (making the UI look pretty). > > Thanks, > Mitch > Email had 1 attachment: > + patches.patch > 23k (text/x-diff) From tpg at ucc.asn.au Mon Mar 30 21:10:56 2015 From: tpg at ucc.asn.au (John Hodge) Date: Mon, 30 Mar 2015 21:10:56 +0800 Subject: [tech] GitLab instance Message-ID: Hello everyone! Since Github is under a little... strain lately (https://github.com/blog/1981-large-scale-ddos-attack-on-github-com) I decided to do some looking into a way of running that sort of service ourselves. The result is GitLab, it's pretty much the same concept as github, but is avaiable to run for yourself. I've configured it to only allow logins using LDAP, so all members can use it (and, sorry, no randoms... unless added by wheel), and it's storing the repos on /services, so they have the same SLA as all UCC data. Feel free to try it out, and forward all complaints to mee/wheel (where we will either fix the problem, or store the message /dev/null) https://gitlab.ucc.asn.au/ John Hodge [TPG] Wheel Member and grumpy, grumpy Midguard