Jump to content

ironbelly

Members
  • Content count

    26
  • Joined

  • Last visited

  • Days Won

    1

ironbelly last won the day on April 12

ironbelly had the most liked content!

Community Reputation

1 Neutral

About ironbelly

  • Rank
    Member

Recent Profile Visitors

301 profile views
  1. Selectively update Workspace

    In perforce this was very simple, we could browse the depot structure on the server and pick and choose which files/folders we wanted to pull down/update. I cannot stress how crucial this is on a large 30 GB+ project being worked on by dozens of people all around the world. We just switched one such project over from Perforce to plastic and to my horror I cannot figure out how to accomplish this and am quite concerned that it isn't possible. The problem is this, we have people in countries like Colombia with slow and sometimes unreliable internet. They are working on 200-300 MB of content on this 30GB project, they only ever need to update or commit from that 200-300MB part of the project and it is literally impossible for them to pull the entire project down in it's entirety because there is absolutely no way that their connection will remain 100% stable for the 35+ hours that it will take to pull it down. Due to the fact that Plastic down not support the resuming of updates or commits any interruption will result in the process being restarted again. We could make them their own branch in theory and then someone closer to the server with a better connection could handle the merge but it would sure be nice if they could simply choose which folder they updated. It's such a common sense feature of perforce that I just assumed it would be in plastic so I am assuming it is possible but I don't know the method of doing it and would appreciate any help here.. thanks
  2. Let's talk about file transfers

    Another option would be not locking the entire folder tree/workspace when committing and allow us to submit multiple commits simultaneously by opening multiple plastic clients. So if I have 1000 files I open one client, select files 1-500 and start committing those.. Then I open a second client, select files 501-1000 and start submitting those at the same time
  3. Let's talk about file transfers

    I'll chime in here and throw my vote behind a resume feature. I have been trying to commit 63GB of data over the last 2 days without avail and really wishing Plastic would be able to resume interrupted transfers right now. To be clear, I'm not splitting this commit up into 2 smaller commits however the reality is still that with remotely distributed teams where members are 4000+ km away from the server you are going to be seeing transfer speeds of 10-30Mbps and when it comes to transferring such large quantities of data at those speeds over those distances things are going to take a lot longer and the chances of being interrupted are going to be a lot higher. Something else that I've brought up in multi-threaded uploads. We currently have the ability to increase download threads which is awesome, but if we could get off this single TCP thread for uploads, at the very least most people would be able to cut their transfer times down by a few orders of magnitude which could be part of the solution for what we're talking about here. Other than that, taking a page out of the FTP handbook and allowing the resumption of interrupted transfers would be glorious
  4. Update Workspace Not Working

    Thanks for re-iterating, this is definitely what's happening over here.. The question is how do I change it back to the dynamic configuration? If I right click where you highlighted in your image I can Set Selector and there I see this in the text field: smartbranch "/main" changeset "1064" I assume setting it back to br "/main" co "/main" Should fix this?
  5. I just want to make sure I understand the right way to go about doing this before investing a bunch of time potentially going on a wild goose chase. Currently we are running a Plastic 6 server on Ubuntu 14.04 linux but if I am to understand the instructions correctly it isn't the plastic server that initiates these triggers at all, it's the client, is that correct? What I want to do is monitor a specific repo, and anytime anyone checks something in, it will update an RSS feed. What is the best way to set this up? Should I install and setup a client on the server, create a read-only user who only can access the repo in question and then create the ruby scripts? What if I want to monitor a couple repos?
  6. Update Workspace Not Working

    I have this same problem right now but to be honest Carlos, your answer above doesn't seem to make much sense, related to the problem at hand. You seem to be describing the normal behavior of the update workspace button. The problem, and the point of this post from my perspective, is to address the issue that it is not working as you've laid out. For example, in my case, I am connected to a repo. It has one branch(main), my workspace is setup to that branch. I am on changeset 94 and the most recent changeset is 95. I go to the items in workspace tab, hit update workspace and it tells me that 'The workspace is up to date @cset:94', which is clearly not true as I can see cset 95 in my find changesets tab. The only way I can pull down cset 95 is to go into the find changesets tab, right click on changeset 95 and click 'switch workspace to this changeset. Only then does it download the updates. This is now the only way I can pull updates down at all on this specific repo, the update workspace button has ceased to work at all.
  7. I might have jumped the gun here in my celebrations.. While I no longer am prompted with the accept cert message everytime I connect I am unable to replicate from one repo to another on the same server. When I go look into the debug logs I see a few things, but this seems to stand out: As well as this: followed by this: and ending in this: I'm confused by the "Only the server administrator can accept a certificate on the server." as my account is the server owner so I'm not sure what more I can do to elevate this account, I coudlnt' see any in the perms.
  8. Can I change the workspace path?

    I don't suppose that editing the C:\Users\USERNAME\AppData\Local\plastic4\plastic.workspaces would do any good would it?
  9. I didn't find any clear documentation on this in particular about this specific case. https://www.plasticscm.com/documentation/administration/plastic-scm-version-control-administrator-guide.shtml#Chapter12:PlasticSCMSSLcertificates Exists however the wording and the reference to .pem, .cer anf .pfx files threw me off a bit as my wildcard cert from Comodo didn't come with any of those files. Anyways figured I would write some instructions for the next person finding themselves in this situation as it was very easy once I figured it out. If your experience was like mine you should have received a .crt .csr .key and maybe a .ca-bundle file which you may have created yourself. What you are missing for plastic SCM is a .pfx file, but fret not as it is very easy to make, just go to the dir with your key and crt files and run the following: openssl pkcs12 -export -out STAR_YOURDOMAIN_COM.pfx -inkey STAR_YOURDOMAIN_COM.key -in STAR_YOURDOMAIN_COM.crt Replacing STAR_YOURDOMAIN_COM with whatever the name of your domain or .key / .crt files are that you've received from your cert auth. That should generate a .pfx file, from there I just copied that file into my /opt/plastiscm5/server dir, and renamed the old ssl-certificate.pfx file to ssl-certificate.pfx.bak just to be safe. After this I went into my /opt/plasticscm5/server dir and opened up remoting.conf for editing, you should be looking for the following line: sslPfxFile="ssl-certificate.pfx" sslPfxFilePassword="|SoC|2ogBDa8GmifTjC7UKp4KuoF0/jWYlXy2" Replace the ssl-certificate.pfx with your file and replace the password with whatever you set when you created the file. If you didnt' specify a password you can have it empty (ie: sslPfxFilePassword="") The only thing I made sure of was that in my /etc/hosts file I had a hostname that linked up with this, ie: scm.mydomain.com That's it, restart the server and you should be off to the races without any more errors.
  10. Ok, problem solved.. Recovery tool worked like a charm.. thanks guys!
  11. I have also concluded that I need to go into the server and try to pull this changeset out of the blobs.dat file. I did a test where I added 10 MB worth of data to the repo, then backed out of that changeset and deleted it and on the server the repo size didn't go down at all.. So this leaves me to understand that even though we lost it locally and what not the files are still sitting there on the server. It simply asked for confirmation, I didn't comprehend the weight of our actions here, specifically switching to the previous changeset causing the local files to be permenantly deleted. My assumption, which is painfully wrong, is that plastic would keep a local cache of these types of things to prevent things like this or things like having to redownload entire branches every time you switch instead of simply loading branches from the cache when you switch. I have come to understand this now, again wrongly assuming that there would be a local cache or these types of things. I agree. At most an archive or hide changeset option would have been better.. Yes, I am using Jet. I've contacted support but have yet to receive a reply so I don't know if they are around.. I'll prep the entire repo for them and send them a link where to download it. I was assuming, seeing as the repo size didnt go down when we deleted this stuff, that the files are in there, I just have no way of extracting them. Is there a tool I can grab to do this?
  12. Thanks for the reply. The situation is that we created a branch as a child of the wrong branch by mistake, added files to that branch and submitted. Realizing our mistake we wanted to delete this branch, and make a new one as the child of main. So the logical course of action was to just delete this branch, make a new one properly and recommit the files. So we went to delete the branch, were confront with the 'can't delete branch with changesets error' message and then went on to delete the changesets. Once deleted we then deleted the branch. After that was done we created the new branch and then went to re-add and re-submit the files and low and behold they were gone. This alone seems insane, as why would it delete the local files when all we did was delete the changesets and the branch, which I would assume at most are server operations. But then I realized that we had to back out of the current change set to delete it, so by going back to the changeset before we added/checkin the files it probably removed them locally in the current workspace. This is very bad as it appears that when doing something like this, or switching workspace plastic isn't storing or caching anything locally at all it is flat out deleting it because our next thought was that there would be NO way that plastic would permanently delete the files locally, worst case scenario it would just send them to the recycling bin. So we went to check there but the files are no where to be found.
  13. We just deleted a branch from plastic and now all of the files that were part of that branch locally are gone.. not in the recycle bin but I am not prepared to believe that plastic would do something as stupid as to permanently delete the local files if someone deletes a branch. Assuming the files are either hidden locally somewhere or on the server, so any help would be appreciated here
  14. Client / Server Version = 6.0.16.866 DB = Jet Anytime I try to access the branch explorer I see an error box saying " There has been an unexpected error " The method or operation is not implemented". For more information check the server log. I can create top level and child branches in the Branch section of the client without any issue but as soon as I go into the Branch Explorer that when that error comes up and I cannot see nor do anything. Here are the server logs: ***DEBUG LOG*** ***ERROR LOG***
  15. Finding and Deleting files/folders from repo

    Answered my own question as I was typing it.. Instead of using the plastic client I went through windows explorer and brought up all of these folders, selected them all, right clicked, went into the plastic context menu and deleted them there. Easy peasy
×