Sourceforge site is currently in Disaster Recovery mode

[Elixer]

The sourceforge.net website is temporarily in static offline mode.
Only a very limited set of project pages are available until the main website returns to service

Been that way for almost 2 days, anyone have any news?

Ever since the new owners came along, seems things have gone downhill.

 

[Jodell88]

It went TITSUP

They had a storage infrastructure fault, according to their twitter.

 

[BarkingGhostar]

I bet it was more of a revenge hack job.

 

[lxskllr]

Sourceforge can go to hell. Hopefully it stays offline. They give free software a bad name.

 

[DaveSimmons]

They've been hijacking projects then wrapping them in scamware for awhile now.

Also "mirroring" popular projects from other sites to get people to download their infested versions instead of the clean ones from the primary sites.

 

[Red Squirrel]

Sourceforge can go to hell. Hopefully it stays offline. They give free software a bad name.

This. I hope with this incident has a side effect of that info becoming more common to people who may not know. Any SF thread in a forum will most likely cause that to be brought up and people will be informed.

 

[Markbnj]

Sourceforge can go to hell. Hopefully it stays offline. They give free software a bad name.

Yeah I wouldn't be surprised if it was an attack of some kind. They really went over to the dark side hard.

 

[ultimatebob]

Pisses me off, because I have an open issue with Clonezilla that I can't get answered because their support forum is down.

Why Clonezilla still uses that shit site for hosting is beyond me.

 

[Elixer]

Supposedly, it was a hardware firmware issue of some kind.

 

[lxskllr]

Supposedly, it was a hardware firmware issue of some kind.

They probably patched it, but Ask toolbar gummed up the works.

 

[John Connor]

Github FTW!

 

[Red Squirrel]

Supposedly, it was a hardware firmware issue of some kind.

SAN firmware can be a pain. We had an Equalogic SAN at a place I worked and the big selling point with these was how it was super easy to just add more enclosures live and everything just converges together. You can make it redundant, or just use more spindles, grow the arrays across the new enclosure etc all live... BUT... and a big BUT, it only worked if the firmware versions matched. Unless you buy two enclosures at the exact same time, of course they're not going to match, the newer one will have a newer firmware. The only way to update the firmware was to take it offline. Who the hell takes storage offline, ever? Well we had to. Totally defeats the whole purpose of having a feature that makes adding a new unit seamless, because it sure as hell is not seamless if the firmware update requires a reboot.

There was also a certain firmware version that if you updated it, it would start to corrupt data. Scary stuff. We got lucky and they were not at that version yet when we updated, but another one of our clients got hit.

I'm going to stick with my 24 bay enclosure running Linux and mdadm raid TYVM. Cheaper, easier to operate, and less points of failure.

 

[Ayrahvon]

Where do people go instead of sourceforge these days? It was my go to for a long time.

 

[ViRGE]

Where do people go instead of sourceforge these days? It was my go to for a long time.

I guess Github? But as an end user (and not a coder) I find Github a lot harder to use.

 

[Elixer]

Where do people go instead of sourceforge these days? It was my go to for a long time.

Sourceforge is still the only option for file hosting, and not have a limit of size.
Github doesn't come anywhere close, neither does bitbucket.

However, both bitbucket & github are better for using hg or git.

 

[matricks]

Bintray is free for open source projects, with no file size limit. They had a limit at 50 MB at one point, but this was removed according to their own Twitter.

 

[dave_the_nerd]

SAN firmware can be a pain. We had an Equalogic SAN at a place I worked and the big selling point with these was how it was super easy to just add more enclosures live and everything just converges together. You can make it redundant, or just use more spindles, grow the arrays across the new enclosure etc all live... BUT... and a big BUT, it only worked if the firmware versions matched. Unless you buy two enclosures at the exact same time, of course they're not going to match, the newer one will have a newer firmware. The only way to update the firmware was to take it offline. Who the hell takes storage offline, ever? Well we had to. Totally defeats the whole purpose of having a feature that makes adding a new unit seamless, because it sure as hell is not seamless if the firmware update requires a reboot.

There was also a certain firmware version that if you updated it, it would start to corrupt data. Scary stuff. We got lucky and they were not at that version yet when we updated, but another one of our clients got hit.

I'm going to stick with my 24 bay enclosure running Linux and mdadm raid TYVM. Cheaper, easier to operate, and less points of failure.

Your Linux/mdadm setup is adequate for low-performance applications. Higher performance SANs will tend to require more stupid-hard-drive-tricks and ridiculous numbers of spindles to keep up high IOPS. When they do that, they become more dependent on the behaviors of the disk firmware.

It takes about a month for our QA people to clear a firmware version for deployment to customer systems, and that's if they don't find anything wrong. It's actually kinda scary how many firmware versions don't meet our standards.

I'd contend that your linux box actually has more points of failure, though - a typical expensive (EMC) SAN box has redundant controllers, redundant SAS channels, etc. It's more things that can break, but fewer things which can take down the entire system when they do. Keeps SAN admins employed though.

I've overheard some pretty crazy stuff, particularly regarding disk firmware. I wonder what storage system Sourceforge is backed by? Is anybody tattling?

There are a lot of firmware updates we can do online now, though, by power-cycling components one at a time. Off-lining the whole system shouldn't be necessary.

 

[Elixer]

Looks like they had bad file corruption going on, so Red Squirrel seems to be right on the money for their issues.

http://sourceforge.net/blog/sourcef...eforge-infrastructure-and-service-restoration

..We responded immediately and confirmed the issue was related to filesystem corruption on our storage platform. This incident impacted all block devices on our Ceph cluster. We consulted with our storage vendor when forming our next steps. We have since been working 24×7 on data restoration, data validation, and service recovery.

 

[John Connor]

Just downloaded PHPMussel from Sourceforge last night with no problem.

 

[saratoga172]

SAN firmware can be a pain. We had an Equalogic SAN at a place I worked and the big selling point with these was how it was super easy to just add more enclosures live and everything just converges together. You can make it redundant, or just use more spindles, grow the arrays across the new enclosure etc all live... BUT... and a big BUT, it only worked if the firmware versions matched. Unless you buy two enclosures at the exact same time, of course they're not going to match, the newer one will have a newer firmware. The only way to update the firmware was to take it offline. Who the hell takes storage offline, ever? Well we had to. Totally defeats the whole purpose of having a feature that makes adding a new unit seamless, because it sure as hell is not seamless if the firmware update requires a reboot.

There was also a certain firmware version that if you updated it, it would start to corrupt data. Scary stuff. We got lucky and they were not at that version yet when we updated, but another one of our clients got hit.

I'm going to stick with my 24 bay enclosure running Linux and mdadm raid TYVM. Cheaper, easier to operate, and less points of failure.

We've got 7 EqualLogic's and I've personally updated each one of them at least twice. Never had to take them offline. They have dual controllers and one controller is updated while the other runs. The SAN fails over to the recently updated controller then the 2nd one is updated. There is a blip in connectivity while that happens...was only a few seconds for us. Never even had to take our production environment down.

 

[Elixer]

Sourceforge can go to hell. Hopefully it stays offline. They give free software a bad name.

Well, seems sourceforge and slashdot are up for sale...so, is this better?

http://arstechnica.com/information-...p-plans-to-sell-off-slashdot-and-sourceforge/

The Company acquired Slashdot Media in 2012 both to provide the Dice business with broader reach into Slashdot's user community base and to extend the Dice business outside North America by engaging with SourceForge's significant international technology user community. The Company, however, has not successfully leveraged the Slashdot user base to further Dice's digital recruitment business; and with the acquisition of The IT Job Board and success of Open Web, the anticipated value to the Company of the SourceForge traffic outside North America has not materialized. The Company now plans to divest the business, as it does not fit within the Company's strategic initiatives and believes the Slashdot Media business will have the opportunity to improve its financial performance under different ownership

 

[norseamd]

So for someone who supports open source but is not involved in the community can you guys tell me all about sourceforge?

Honestly I have used sourceforge for more than a few downloads so I would certainly be very interested in knowing all the bullshit that might be going on right now.

 

[Kaido]

Your Linux/mdadm setup is adequate for low-performance applications. Higher performance SANs will tend to require more stupid-hard-drive-tricks and ridiculous numbers of spindles to keep up high IOPS. When they do that, they become more dependent on the behaviors of the disk firmware.

It takes about a month for our QA people to clear a firmware version for deployment to customer systems, and that's if they don't find anything wrong. It's actually kinda scary how many firmware versions don't meet our standards.

I'd contend that your linux box actually has more points of failure, though - a typical expensive (EMC) SAN box has redundant controllers, redundant SAS channels, etc. It's more things that can break, but fewer things which can take down the entire system when they do. Keeps SAN admins employed though.

I've overheard some pretty crazy stuff, particularly regarding disk firmware. I wonder what storage system Sourceforge is backed by? Is anybody tattling?

There are a lot of firmware updates we can do online now, though, by power-cycling components one at a time. Off-lining the whole system shouldn't be necessary.

It's crazy how niche you can get with storage...I'm kind of glad I don't work for larger companies with crazy storage needs like that. I've recently started using Synology NAS setups for business purposes, which I'm a lot more confident in than say Drobo. Pretty nice for the money...iSCSI for Windows Servers & VMware, active & passive mirrors if you use a pair of units, integrated backup if you throw in a third unit (and the software setup for mirroring & backup is completely painless, all built-in turnkey stuff). You can build a 60TB NAS with 12TB parity (using SHR-2), mirroring & backup for about $25k, which is pretty neat for what amounts to maybe an hour's worth of work to pop some off-the-shelf drives in & update it to the latest firmware.

 

[Kaido]

So for someone whosupports open source but is not involved in the community can you guys tell me all about sourceforge?

They were pretty a pretty awesome open-source project repository for a long time, but then started essentially sneaking in adware, Download.com-style (confusing buttons etc.), which really goes against the whole concept & community of open-source projects. Per Wikipedia, "On June 1, 2015, SourceForge claimed that they stopped coupling 'third party offers' with unmaintained SourceForge projects."

Then their SAN crashed.

 

[Anubis]

chrome is the worst piece of adware on the internet