This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.
Messages - alberthrocks
Pages: 1 ... 6 7 [8] 9 10 ... 55
106
« on: March 20, 2014, 01:50:28 pm »
On the topic of Axe documentation... We used to have an Axe wiki, where this kind of stuff (tutorials and “better” documentation) can be found. Mediawiki needed to be updated 24/7 though, and things got messy so fast (spambot fiesta!) that it was eventually shut down. (See: http://www.omnimaga.org/news/axe-parser-wiki-opens-its-doors/msg220368/#msg220368) I don’t think we’ll ever put up another wiki again (unless it’s some other wiki software, NOT Mediawiki). But if you’re really interested in improving documentation, you can either modify the current one (that one HTML file), or get your hands messy by creating another one! Sphinx ( http://sphinx-doc.org/) is a documentation format that creates pretty documentation from simple markup. LOTS of projects use it, and although it is a little overkill for a project like Axe, you can certainly use it to provide meatier details about Axe, including tutorials and the likes. I’d love to see a https://axe.readthedocs.org/ one day! EDIT: Looks like thep ninja'd me!
107
« on: March 04, 2014, 09:39:38 pm »
Hey guys! This is alberthrocks, your friendly invisible CoT! A couple things to clarify: - This is technically Omnimaga v2.0, big upgrade. No kidding, we did a super big upgrade, which is no small feat!
- That said, the upgrade is a WIP. Call it Omnimaga v2.0 SUPER ALPHA ALPHA ALPHA. Consider the website in alpha right now, with tons of bugs and problems at the moment. We're actively working to fix all of them, but for now, we ask that you bear with us during the upgrade. To repeat, the upgrade is not complete yet!
- The old theme WILL be back - we've planned this from the start. This is just a temporary theme that we'll continue to tweak to make better until upgrades are complete. Once done, we'll begin working on porting the older theme to v2.0, and bring it back eventually. You'll have the option to pick the current new theme or the old theme once the upgrade is done.
- We upgraded to protect you from hacking! The reason why we are upgrading this fast is because we were running on an older, unsupported version of SMF. For you guys, that means spambots, super prone to hacking attempts, etc. (Some bots even managed to start going through accounts and steal personal info, like your email!) We upgraded to stop that from happening, as we believe that's really important to protect!
- Help us help you! Please report issues using this form, not here! We want to make this upgrade a success, so please report issues there! Reporting it on that form helps us more than posting about it here!
- Finally, please stay put as we work through this - don't harass/complain/etc. until we announce that we're done! We know there are issues, so please use this form to report it! We're working around the clock to fix stuff, and we'd appreciate if you tell us about the problem! Once we get out of alpha, we'll open up for changes and such. For now, to keep our admins sane, please use the form to tell us about issues and hold tight till its done!
Thanks for reading!-alberthrocks
108
« on: March 04, 2014, 09:23:00 am »
Looks like we had a HARD 2 hour downtime (with the VPS actually being turned off) - sorry for the inconvenience! (This may be due to something on the VPS provider's side, dunno...)
This means that any bots or things that were running before have been turned off, and you may need to start them up again.
ZNC was disconnected but has started up again.
Again, sorry for the inconvenience this may have caused!
109
« on: March 03, 2014, 05:15:27 pm »
Just a quick status update:Current Status:- Status last changed: 3/2/14 (this is NOT the post update time, just the time I last worked on the migration!)
- Time spent: ~6 hours.
- ETA: ∞ - ~6 hours.
- Next step: Small scale local testing.
- TODO: Large scale local testing, code integration, large scale production testing, deployment, and completion of backup.
- Further TODO: Copy data backup to new server, re-do system backup, write system backup restorer, write data backup restorer, run data backup restorer, run system backup restorer, start ZNC on new server, update DNS, do "user migration period", move ZNC config, terminate old server
Completed tasks:- Data program - Completely rewrote logic for data backup. Incomplete. [2/12]
- Data program - Continued to rewrite logic for data backup. Fixed some bugs here and there, still needs another look. [2/17]
- Data program - Finalized logic for data backup. Still untested, but it should be ready for integration. [2/19]
- Data program - Wrote verification code to ensure that complete backups are made. [3/1-3/2]
Progress Bar:[███ ] 30%
_^^^^^^^^^^
111
« on: February 24, 2014, 10:23:03 pm »
TL;DR: really sorry for delays, backup+migration will get done, I promise! Hey everyone! Let's see, where do I start.... I'll start by saying that I'm a horrible server/VPS/network admin, who promised you guys a migration to a faster server and gave... nothing. And on top of that, this admin locked everyone out of their accounts!  For that, I'm really really really really really really really really really really really really really really...sorry for the inconvenience. And even with that, 1000 + ∞ (yes, you can count them, 1000 is possible on the internet) really sorries don't make up much for it. But I promise you that I'll get to the migration.... eventually. If you depend on withgusto for services other than ZNC and require login, please let me know and I will be happy to unlock your account. This includes any special requests, like software installations, server setup, etc.As for reasons/excuses why it hasn't gone anywhere:- The data backup software is 100% custom written. It's written by hand for many reasons, tl;dr (1) partition backup wouldn't really be fun on a VPS; (2) want to preserve current setup as much as possible while still moving to a new server AND newer OS; (3) much more efficient while ensuring that it's a good, solid, safe backup.
- The package backup portion is 100% successful! However... the data backup portion (which I had mistakenly thought would be easy peasy to write) failed miserably. (tl;dr it forgot to backup files in most of the directories - i.e. it backed up /dir/subdir, but not /dir/file, even if it backed up /dir/subdir/file or /file)
- As a result, the thought of it failing and the fact that it needed a really serious rewrite significantly killed motivation. Not sure if this happens to you or not, but project development failure can really ruin your motivation.
- Motivation levels have picked up (still low ish but more like low-medium area), but I just had a midterm last week, and will have 2 more this week and another one next week. Sorry, college > all - gotta make sure I do well on these exams!
- Developers are notoriously bad for underestimating development time - myself included.
No worries - I'm personally motivated to make withgusto the best for its users (thank you for your support)! And there's monetary motivation, too - I just paid around $23 worth of server fees, and have another $15 invoice due 3/1. Eeeek! It'll get done... eventually. I don't want to give any timelines, but I can promise you it WILL get done. Eventually. To prove that, I'll give you guys the current status on the move. It'll get updated on the IRC EFNet/OmniNET channel status of #IRP, maybe here if I get time/motivation. Current Status:- Status last changed: 3/2/14 (this is NOT the post update time, just the time I last worked on the migration!)
- Time spent: ~6 hours.
- ETA: ∞ - ~6 hours.
- Next step: Small scale local testing.
- TODO: Large scale local testing, code integration, large scale production testing, deployment, and completion of backup.
- Further TODO: Copy data backup to new server, re-do system backup, write system backup restorer, write data backup restorer, run data backup restorer, run system backup restorer, start ZNC on new server, update DNS, do "user migration period", move ZNC config, terminate old server
- Completed tasks:
- Data program - Completely rewrote logic for data backup. Incomplete. [2/12]
- Data program - Continued to rewrite logic for data backup. Fixed some bugs here and there, still needs another look. [2/17]
- Data program - Finalized logic for data backup. Still untested, but it should be ready for integration. [2/19]
- Data program - Wrote verification code to ensure that complete backups are made. [3/1-3/2]
Apologies again for the delays! EDIT #1: Minor edits here and there. Please take note of the stuff in bold!
112
« on: February 24, 2014, 02:26:24 pm »
Wow, it's been a while. Happy Decennial! 
113
« on: February 22, 2014, 01:01:36 pm »
I've never had this happen. Did you google it to see if other people have the same issue?
I wish it were that easy, but I only get questionable answers that don't really help. (Examples include this and this, if not much more.) If, as a result of the switching your IP changes, and the server checks your IP, then there is no way to overgo that.
Nope, Chrome still keeps me logged in. And it's not because of Google bias towards Google-y sites (3 tabs of GMail, 1 Google Voice, 1 Google Calendar, 1 MightyText <-> Google Account), but even other websites like Facebook, Twitter, Hotmail, etc. are logged in. This is an agonizing problem that really affects my use of Firefox - believe me, after logging back in for 5+ sites (not mentioning the chaining of 3 Google Account logins), it gets annoying. I can live with the memory hogging Chrome if it means that I don't have to worry about getting logged out (and having to spend some time logging back in again).
114
« on: February 22, 2014, 12:36:14 am »
Random question - I recently switched to Chrome because of an annoying problem, but kinda wanna switch back to Firefox.
The problem? Whenever I switch WiFi networks, everything logs out for no apparent reason. Yeah, it's "more secure", but it's really annoying, especially if it's just a switch between the same network. Is there a setting that can be changed to prevent this? This behavior does not happen in Chrome.
115
« on: February 11, 2014, 01:30:19 am »
Did you upload this to ticalc.org? This is gonna be big! (And link profusely to omnimaga.org to generate tons of new users )
116
« on: January 28, 2014, 10:44:52 pm »
The global data freeze is now in effect - you will no longer be able to log in. You can still SFTP data back and forth, but we recommend not changing any data on the server, as it could mess up the data backup and result in your data being lost! (After the data backup is finished, we will NOT copy changes you make to the new server, even when unlocked! It is now YOUR responsibility to make a backup of any changes!) If you REALLY need access to the server, let me know and I'll unlock your account. In the mean time, you can login to your account and see this live status screen! 
117
« on: January 23, 2014, 12:21:34 pm »
As always, you are always welcome at withgusto!  We're moving stuff right now, so there might be a delay, but otherwise I'm happy to set you up! (aka: my verbal commitment from yesterday still stands!)
118
« on: January 21, 2014, 01:19:21 am »
This VPS has been leased to Juju. Any further inquiries about the VPS should go to Juju, as he has already paid for 2-3 months.
119
« on: January 17, 2014, 03:23:05 am »
Server Update - 01/17/2014
Importance: IMPORTANT
Author: albert
Date: Fri, Jan 17th, 2014 at 12:18:59 AM PST
Blog post: http://withg.org/blog/index.php?post=1389946739
withgusto Migration Update #1
==============================
We have reached Day 5 of migration! Writing status updates are painful
(especially since we want to be on schedule), but I didn't want to
have too much panic, so I'll be "open and transparent" (BUZZWORD)
about what's going on behind the scenes, and some plans ahead!
Current Status
---------------
I'm writing the backup software in-house. Why? There's no software out
there that does the backup that I'd like to do - a very fine, intensive
but safe backup. (I can't do partitioning because that would be a
little... weird to copy over to another server. And it's really big,
too. And it can only result in a clumsy mess... at least for VPSes!)
Due to some unforeseen bugs within Python-APT (Debian's Python interface
to their APT packaging system), development of the tool is taking a
bit longer than usual. Development time was slated for 1-2 days, but
it has gotten a significant delay. However, the backup itself shouldn't
take too long as long as the number of packages are low (<1500),
internet is fast and speedy, and there's enough disk space.
I just finished writing the workaround for that bug today, so normal
development should resume tomorrow, and we could be seeing system backup
start as early as tomorrow evening!
I won't go into too much details - we'll save the fun technical stuff
for after the migration!  Nevertheless, for those who are curious,
here's a link to the workaround code:
https://gist.github.com/alberthrocks/4bef423e8ecc4df357ce
Moving Forward
---------------
The Global Data Freeze will still be occuring today,
Friday, January 17th, 2014. As for time, the freeze will occur at
11:59:59 PM on that day, so there should be plenty of time to move and
sort things out! Again, we ask that if you have any large data files,
please download them to your computer for backup and delete them from
the server!
We switched to CloudFlare for faster DNS propogation, so hopefully
when the time comes, "flipping the switch" will not take too long! We do
not anticipate any web service outages at this time.
ZNC service will be switched over sometime next Tuesday. We'll
provide another update as to what time. To prepare, make sure you have
a backup IRC client/user ready if you need to keep ops!
OmniNET IRC on this server (irp.irc.omnimaga.org) will be going down
soon. (No worries, it WILL come back, but only after the new server is
up and ready!) If you are on OmniNET, just switch to irc.omnimaga.org.
If you are already there, you should be fine. Note that NickServ and
ChanServ data will be wiped on the new servers that are supporting
OmniNET, so you will have to reregister. Sorry for the inconvenience!
120
« on: January 13, 2014, 02:25:12 pm »
Server Update - 1/13/2014
Importance: IMPORTANT
Author: albert
Date: Mon, Jan 13 11:25:00 AM PDT 2014
Blog post: http://withg.org/blog/index.php?post=1389641100
withgusto Update: We're moving!
======================================================================
Welcome to the new year! You've heard about it, and now it's happening!
As part of our consolidation plan, we are shutting down a few servers and
moving to a cheaper, better server!
Moving, moving, moving!
------------------------
As part of the moving process, we would really appeciate if you could do
the following:
* If you have any big files on the server, please download them to your
computer so that we can pack and move lightly to our new server.
As always, make sure to md5sum/sha1sum the file on your computer
BEFORE deleting it from the server!
* If you have any running processes that are necessary, download their
files now and be ready to move them to the new server.
* Also, if you have any running processes, make sure to start them now
and keep them running with an "always-on" script. (Usually, a while
loop running the program will do, WITH a time delay. If you don't
have a time delay, the program will use up all of the CPU, backups
will slow down, and your program WILL BE TERMINATED! So don't do it!)
Our moving process will proceed in five stages:
* System freeze and backup (1/15/2014-1/18/2014) - We will "freeze"
the system and prevent any changes from being made to the system files.
You can still login and do things on the server at this time, but you
will NOT be able to install anything on the server. We will then proceed
to back up only the system files. Within the same time period, once the
backups are complete, we will transfer and restore them to the new
server, starting on 1/17/2014.
* Data freeze and backup (1/17/2014-1/20/2014) - We will "freeze"
the entire server (including user files) and prevent any changes from
being made to the user directories. You will be LOCKED OUT during this
phase, and will be denied login access. At this phase, your shell
will be replaced with a status page on the backup and moving of the
server(s). If absolutely necessary, we may make exemptions on a case
by case basis, but keep in mind that changes you make will NOT be
backed up. Like the previous stage, files are backed up and being
restored to the new server at the same time.
* Verification day (1/20/2014) - This is the day when the server will
be "unlocked". You will be allowed to login again on this day. Check
your data on this day to make sure everything has arrived safely.
If you have any issues on this day, please let the admin know ASAP.
After this day, the server will be DELETED! Alternatively, if it
is just a minor problem with some files, simply copy the files from
this server to the new server.
* Old VPS/Backup Service Termination (1/21/2014) backup2.withg.org,
the old withg.org server (this one), and the backup service,
JungleDisk, will be terminated on this day. (JungleDisk made daily
backups of the main server... but costed too much! Backups will be
done in the future in a decentralized environment... and as always,
it is up to the user to backup as well.)
* New VPS restoration complete (1/22/2014) withg.org will now point
to the new server, and services should be restored.
Keep in mind that backup2.withg.org and withg.org (this server)
will be TERMINATED. All necessary preparations should be done by
1/17/2014 (or earlier). The freeze will prevent you from logging into
either server, so be ready!
Servers backup2.withg.org and withg.org will be moved to the new
server. Data should be intact, but you should make sure to check on
verification day, 1/20/2014!
A final SQL backup will be made right before verification day. We will
perform this last due to the ongoing nature of SQL.
A graphical timetable can be found here:
Services
---------
Web services will continue to run for the duration of the move, and should
not be affected. DNS will be changed immediately after the move has been
completed for the system and the web files, so DNS propogation hopefully
won't be an issue.
If you are having problems accessing withg, we will have a new site pointing
to the new server, v2.withg.org.
SFTP will still continue to run during the data freeze period, just not SSH.
You may continue to make your own backups until the servers are terminated.
ZNC will have a brief interruption on verification day, so make sure to
start any additional IRC clients on your side if you need to keep ops.
Any background processes will be terminated when the servers terminate, so
make sure you start them up ASAP!
New Server Funding
-------------------
This is important! Starting today, all funding for the server
will be 100% user-donated. The owner will no longer provide any funding
due to financial constraints.
Our new provider has these plans:
* $4/month: 1 vCPU, 512 MB RAM, 15 GB SSD, 1 TB bandwidth
* $8/month: 1 vCPU, 1 GB RAM, 25 GB SSD, 2 TB bandwidth
* $16/month: 2 vCPUs, 2 GB RAM, 50 GB SDD, 3 TB bandwidth
* $24/month: 2 vCPUs, 3 GB RAM, 75 GB SSD, 3 TB bandwidth
* $32/month: 4 vCPUs, 8 GB RAM, 200 GB SSD, 5 TB bandwidth
Why would I show you these plans? Simple - the donations that are
received each month will determine which server specs withg will have.
Starting today until the new VPS is purchased (1/17/2014), donations
for this month are being accepted. If the community wants a super speedy
server for $32/month, pitch in $32/month! If the community wants a
decent (but not too crazy) server, pitch in $24/month! A lesser server,
$16/month, and so on... According to the survey, our ideal server is at
$16-24/month range, and if everyone pitches in, we'll have extra money
left over for the next month! (We apparently have a LOT of willing donors...
BUT don't take that as an excuse to not donate! If you don't donate, chances
are that everyone else won't donate, either...)
Note that all the donations MUST be received by 1/17/2014. Even if you
donate a day after 1/17/2014, funds will be allocated for next month,
not the current month!
This provider is also pretty great in terms of flexibilty! If money is
short, but we have enough for a lower plan, I can easily downgrade to a
lower plan and things will work out perfectly! (Not sure about downtime,
but from what I've heard it is minimal.)
I truly apologize for sounding harsh (like an angry landlord, no less!),
but I really need to cut back on spending, as money is short for both me
and my family.
That said, please donate!
http://withg.org/donate/
On a more positive note, the plans are MUCH cheaper, and you guys get
really fast SSD storage, so that's something to look forward to!
The Future (again)!
--------------------
There are some exciting announcements that will be made in the future,
so stay tuned!
Thanks again for flying withgusto! ✈
We'll have a brief stopover on an island, and we'll be transferring to
a new flight, withgusto v2!
Your captain (aka admin),
Albert H (alberthrocks)
Pages: 1 ... 6 7 [8] 9 10 ... 55
|
Show Posts