AdamS

I'm not sure if I'm using the correct terminology here but I think there may be a possible memory leak with 1.8.3..? I'm running a case at the moment that keeps crashing (intella just disappears, or windows freezes completely) and the memory usage for one of the java.exe instances is up to 60GB (64 GB memory available). To be fair I've not looked at the memory usage before as I've not had any crashes, this will be about the 4th case I've indexed since the upgrade to 1.8.3 but this is by far the largest data set and probably has the most variables when considering the amount of sub folders involved. Happy to send log files if this latest index crashes again. Edit : possibly Eraser was causing the crash, uninstalled and indexing again, however memory is still pushing upwards of 60GB, but no crash yet...

With the backup process would it be possible to expand the functionality of this slightly, to either take on an incremental approach, so it's only backing up the files that have changed since the last backup, or have the ability to dump the backups into some sort of folder structure for each case. By that I mean lets say I have a case called "client A", when I back that case up for the first time after I select the backup location Intealla automatically creates a simple folder structure: "Client A" subfolder "230215_1336" and dumps the backup in that folder. Next time I backup the same case Intella checks the backup location for a parent folder the same name as the case, Client A in this case, seeing the folder already exists it creates a new sub folder with the date/time and dumps the backup there. That way we will have the ability to go back in time as it were to earlier stages of the case.

I'm seeing quite a few emails on a case I'm working on now where the actual email address is not being displayed in the header or anywhere, it's simply showing up as "Robyn"<Robyn> as it would appear when the email address is already in the contact list. While in many cases it's obvious who the email is coming from due to signatures this is not always the case. Is this information that is not included in the header under some circumstances? I'm thinking when emailing within a domain where all users are on a global address book would Outlook strip the full email address from the header as it's not needed? And if that's the case the receiving email within the domain still has the full email address populated, so that sort of confuses me.

I was playing around with the latest Connect and I created a new account type called 'Case Manager', I gave that account the permission 'Manage own case and authorize reviewers'. I then logged in as 'admin' on the Connect server, authorized my own account as a 'Case Manager' and shared the case. I was anticipating when I logged in to connect there would be a new tab or new options on the dashboard to enable me to nominate reviewers but I can't see anything different from that log in to my normal reviewer log in. Is this a glimpse of a feature yet to come or am I missing something obvious? Edit: actually on further experimenting perhaps I was misunderstanding the role of this permission. I was thinking in terms of access control where I as the admin could create a case and then designate a 'case manager' in my clients office. Then I share that case from my server and then the client can log in as a case manager and then internally create reviewers who could log in and review the data. But it appears this is more for an internal thing from what I've figured out, so I could set up a 'admin light' profile and allow other people here to set up and administer individual cases while I as the main admin could over see any/all cases on the server.....is that right?

Have just had a look at 1.8.3 and have to say I'm really liking the way Connect is developing. A very small thing, the ability to add cases independently of TEAM is great, but would it be possible to have a 'browse' button to enable us lazy people to navigate to the case.xml file we want? At the moment we have to manually type in the address to find the file. A very minor thing but just a bit more user friendly if we can navigate to the file directly. Oh and any word on the ability to merge cases?

You may have to build the search in stages rather than one all encompassing search term. Maybe something like: Search for Sender/From is "@acme.com" Then within that set where To/CC/BCC is "@acme.com" Then sort in the preview pane by the "To" field and disregard any emails that include external emails addresses (if that's your goal) Tag the remainder as internal emails Rinse and repeat for the "@acme-cfg.com" emails but use the date filter to isolate the relevant time period. It's not as elegant as a single search term however experience with Intella has shown me that sometimes it's safer to build the search in this manner to ensure nothing is missed through any misinterpretation of the intended search parameters.

I know ScanPST doesn't always fix all the errors with a single run. As a general rule run Scan PST over and over until you get a 'no errors located' message, this can sometimes take 6 or more runs. There is a batch version of ScanPST somewhere that you can set it to auto run until it's error free. Other than that as a general rule I run Scan PST on all PST/OST files before I ingest them with Intella, more time on the front end can save bucket loads on the back

Have you considered using a different tool for the remote access? Team Viewer for example? I know RDP is there and it's free but Team Viewer would give you far greater functionality and may resolve the issue instantly without having to wait for a possible fix.

I know that question was directed at Phil but for comparison purpose I have no mapped shares and am running on a local machine.

I've never had that error on any version of Intella Phil, strange. There has always been a delay populating the file structure but usually no more than 5 seconds. Edit : I'll have to correct myself here as it seems my memory after a months leave was a bit generous, file structure took around a minute to display on a case I just created. From memory you are using all remote work stations aren't you? Does this error occur when you are directly accessing the host machine?

2GB ram on reviewers running Win7 is severely limiting as well...not sure that this would equate to such a long delay opening documents but it may be contributing.

That's pretty much my process only I don't bother compressing and you don't need the ICF file if you are keeping the case data. The amount of time it takes to compress large data sets vs how cheap hard drives are, I just buy more drives for storage. I tend to rotate and delete historic cases once they are about 2 years old but that will depend on your own policies and needs I guess. If you want the ICF file then you just need the original pre-processed data, this would allow you to reindex the data, import the ICF and be back where you started.

Yep spot on, use the column filters to sort and you will easily be able to spot the emails with only one recipient, that way you have emails that only involve the two desired email addresses.

Use the search filters to search only within the 'sender' and 'from' fields with one email address, then search in the 'To' field only with the other email address. The results that intersect in the balls should be emails sent from one user to the other. Rinse and repeat changing the sender and receiver email addresses to find the emails going the other way.

Hi Christiaan, wouldn't the size of all the documents already be known due to the indexing process? I wouldn't have thought anything would need to be calculated here.

No I don't think it's designed that way, you will need to take a slightly different approach when dealing with child/parent items depending what you want to tag. My usual process is to first examine only emails (parent items) that way I will capture any child items as part of that tagging process. Then once I've completed that I will go and view any other files which may not have parent items separately.

Tag the parent item rather than the attachment only. You can either change the tagging behaviour in settings so that tagged items automatically include parent items, or my preferred method is once I'm finished tagging I highlight all tagged items that are child items, right click and select show parents. Then apply the tag to the parent items and remove the tag from the child items. Then export as normal and you will have parent items with the attachments intact.

As far as I know there is no automated "undo" or "redo" functionality like with MS Word, and on examining the logs it appears the only information that is logged is the fact that tags are applied or removed and from what I can gather the hash for the name of the tag is recorded but nothing else. Which makes me think of a couple of questions for our Intella people: Why list the tag hash as opposed to the tag itself in the log file? Thoughts on logging the tagging process in a more granular fashion to allow manual recreation under these circumstances, perhaps even as an optional logging option. All that would be required would be the Item ID and tag applied along with the date/time. Even a large case isn't likely to have so many tags that this will blow out the size of the log file much larger than it already is.

Christian I think the preview reference is when the preview pane is active in Outlook and the email itself can largely be read without actually having to 'open' the email.

I would think in this case possibly not as the issue (at least in my case) was caused by some corruption to the case data. Having said that it wouldn't hurt to back it up in case you are worried, but a full re-index should fix the issue with no working data being lost. With the latest update indexing is much faster so hopefully the down time won't be too much.

Gary I have seen this quite a bit on a case that became corrupted due to my own impatience I cancelled the indexing process and got sick of waiting for Intella to finish cancelling, so I killed the process in task manager. Any time the case was opened after that the exact symptoms you describe above occurred, and in addition sometimes when viewing actual emails or tags the preview pane content would be blank as well. The only fix was reindex the content. This can be done without losing any tags (or at least I have never lost tags from re-indexing content).

dk_mitch as far as I'm aware two of the better OCR packages out there are Abby and Maestro (I prefer Maestro), however both are very expensive and given you are only talking about a few hundred documents I think you'd be better off looking at Adobe's base products or even Abby Finereader. Of course Maestro Recognition server offers a free trial that will handle up to a few thousand pages so this may actually do what you want and also give you a good look at the software.

mark you can create a work file from the old case (tags etc) then export that. Reindex the old content along with the new sources, import the work file and you will have captured all the work done previously. I know it's a bit of time however the new indexing engine in 1.8 is much faster and you should be back to where you need to be within a day. I recently did this with around 400GB of mail archives and was back up and running the following day.

Petruf I think this is possibly due to the conversion of HTML (or other) encoded emails to PDF, as far as I'm aware the formatting will always be lost in this conversion, plus the PDF view gives technical information as well such as headers etc. I agree the layout is not easy on the eye and if you have an email with lots of connected FWd's and CC's etc it looks very messy. I'm not aware of any way to export the data to PDF yet maintain the formatting and style of the original email, I suspect the only way this could be accomplished would be to export to JPG or some sort of picture file which is not currently an option. You can cut down on the clutter by reducing the technical information that is included in the PDF export (headers, comments etc) but you will still lose the formatting and style. I would put my support behind a request to have the ability to export to tiff or jpg (or any other format, possibly HTML) which would allow us to preserve the style and formatting of the original email yet have it in a universal 'safe to use' format. While technical people can make sense of the PDF format non techies (lawyers, clients etc) will get a headache so if we can give them something pretty and familiar it would be fantastic. The only downside I can see to this is how to deal with embedded attachments. And while I'm here a quick question / request. As part of the indexing process Intella appears to treat small graphic files from signatures (intella calls them embedded attachments) as normal attachments, so after indexing the details pane will list all the little .png files as attachments to the email, when in reality they are part of the body. This gives an unrealistic impression when attempting to find only emails with attachments. Is there any way to have Intells differentiate between embedded attachments and normal attachments? I forsee an issue possibly if the attachment is actually embedded inline as some pictures can be, but even having a separate field for embedded attachments and normal attachments would be a great assistance until the software can tell the difference completely.

Ah yes IWR indeed. I just finished indexing the consolidated data set 408 GB of NSF archives and a few hundred MB of documents (PDF, .doc etc). As a point of interest the same set of data took several days to index previously, it took around 24 hours this time, so massive improvement there. I suspect NSF archives are slower to index in general from what I've seen. I'm importing the IWR files from the other 3 matters now so I will let you know how it goes.