Jump to content

jayivan

Members
  • Content Count

    33
  • Joined

  • Last visited

  • Days Won

    1

jayivan last won the day on November 3 2016

jayivan had the most liked content!

Community Reputation

1 Neutral

About jayivan

  • Rank
    jay ivan

Profile Information

  • Title
    senior consultant
  • Gender
    Male
  • Location
    Brooklyn and New York
  • Interests
    Current Active Projects:
    - Artist Foundation's Catalog Raisonne, Collections Management, and Inventory Tracking
    - Museum Artwork Catalog and Library Catalog
    - Music Label Asset Management, Tour Management, and Office Admin including integration with multiple websites via MySQL ODBC and Google Calendar (Zulu)
    - Nonprofit Membership and Publication Database: membership and dues management, and publication management (using tagged exports to InDesign)
    - Art Appraisal Workflow management (tracking of works and building of complex appraisals)

Contact Methods

  • Website URL
    http://macktez.com

FileMaker Experience

  • Skill Level
    Intermediate
  • FM Application
    17

Platform Environment

  • OS Platform
    Mac
  • OS Version
    High Sierra

FileMaker Partner

  • Certification
    Not Certified
  • Membership
    FileMaker Business Alliance

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

  1. The SSL mismatch and Interactive Containers errors in macOS FMP bug persists. FMP/FMPA macOS clients (v15 and forward I believe) using bonjour browsing to connect to database will see ERROR when trying to play audio files in interactive containers (or cannot play videos or work with PDFs). Connecting via FQDN does not have this problem. But why should local clients have to go out through the WAN and back in? FileMaker Go clients and Web Direct clients will be warned about mismatch but can proceed and interact with containers. Windows FMP/A clients are fine as well. Only macOS FMP/A clients have issue. This has been around a while...any possibility of a fix? (Modifying /etc/hosts on every computer or setting up an internal DNS server are not viable options in some client environments.)
  2. FileMaker emailed an alert to cloud customers warning that they may not be able to login to their FM Cloud console using valid credentials. My cloud 17 client did not have a problem logging in, but my cloud 16 client did. WARNING: Their support team is requesting administrators to send an UNENCRYPTED EMAIL containing the private key file to resolve this issue. But, they do have an FTP-TLS method for delivering this file. Make sure to ask for this info if you are also locked out of your server. Now, I'd rather not send them the private key file period since it is no longer private, but there seemed to be no option. Any idea what is going on and why they are requesting this unusual step?
  3. I was looking for a tool like this to understand how my FM changes impacted the call being made. Thanks for this! I've never used it and having a bit of a learning curve, but I'll keep plugging away at it. If I can get the tool working, it sounds like it will be invaluable as I expand my solutions to involve more and more API integrations.
  4. I am working on an API integration from FileMaker 17 to Emma. Generally when I work on a new API, I get the curl working on the command line and then translate that into FM. But I've hit a roadblock with one particular API call while my other API calls to Emma are successful. (In the example below, I substituted out sensitive data with all caps.) From the command line, the code is: curl -u USER:PASS -H "Accept:application/json" -H "Content-type: application/json" -X POST -d '{"fields": {"name_first": "TESTFIRST","name_last": "TESTLAST","eligible": "false","expiration": "@D:2019-06-30"},"group_ids": ["GROUPIDNUMBER"],"email": "EMAIL@EMAIL.COM","opt_in_confirmation": "false"}' "https://api.e2ma.net/ACCOUNTNUMBER/members/signup" This command is successful. The server responds with the individuals' emma id and group confirmation. This fails using Insert from URL within FM though. While eventually I'll use FileMaker fields for an individual record's names, email, and expiration dates, I did not use fields in my testing in order to duplicate the command line exactly. In Insert from URL: The URL is "https://api.e2ma.net/ACCOUNTNUMBER/members/signup" The curl option is "-u USER:PASS -H \"Accept:application/json\" -H \"Content-type: application/json\" -X POST -d '{\"fields\": {\"name_first\": \"TESTFIRST\",\"name_last\": \"TESTLAST\",\"eligible\": \"false\",\"expiration\": \"@D:2101-06-30\"},\"group_ids\": [\"GROUPIDNUMBER\"],\"email\": \"EMAIL@EMAIL.COM\",\"opt_in_confirmation\": \"false\"}'" Can anyone identify the problem or suggest a workaround? My experience with API integration with FM is limited, but anything I could get working via the command line, I've been able to get working within FM until this.
  5. I have solutions that include a local file residing on a workstation that interacts with a hosted file. The local file does not need the hosted file for data entry; the connection is to download a job list and upload time entries. In FileMaker 16 and earlier, should the local file had been connected to the hosted file and there was a network interruption, the hosted file would reconnect or fail to do so, but the Local file would not be impacted. In FileMaker 17, if the connection cannot be re-established to the hosted file, the local file is now forced to close. That in and of itself would not be a big deal, but when the local file closes, NO DATA IS COMMITTED and DATA IS LOST. Here's a scenario: a user downloads the job list and begins to do their timesheet when their network connection goes down. So, there's a series of lines entered in the portal when the network connection is down. In FM16 and earlier, the user could keep working or close the local file, and any work in progress would be saved with the local file. The time entered in the portal would be saved because the user would close the file, committing the data. In FM17, when the connection is lost and FM closed the Local file, it doesn't commit the data and data is lost. This change in behavior resulting in data loss seems like a bug. But if this behavior is intentional, I'll need to change the workflow to add commit steps after leaving portal rows, or adding onTimer commits scripts, or some other workaround to protect data and hopefully not interrupt user entry much. Any thoughts on this situation?
  6. After banging my head for a few hours, I figured it out shortly after posting lol. So for those who also missed it... when setting a daily backup schedule, change the dropdown underneath Time from "At" to "Between" and the hourly/minute repetition options appear.
  7. In recent FMS versions before FMS17, you could create a daily backup schedule that would repeat every X hours. I need that in FMS17, but those options are no longer available in the web admin, and I have yet to find a way to do this via the CLI. If I want to run a schedule every hour of every day, do I need to create 24 backup schedules like back in the day?
  8. UPDATE/RESOLVED 360Works support walked me through using the 360Works Admin.jar tool to stop and start the MirrorSync server to resolve this error. So, for the external server address, it seems to be fully possible to use a domain name whose public IP doesn't match the public IP of the router, and to ignore the IP-mismatch warning during configuration setup. But, you may need to restart MirrorSync to avoid errors caused by cached data. Unrelated to this, if you update the internal IP of the hosting server, or update the filename of the database, you may need to make adjustments manually in the external data sources to match those changes.
  9. I was able to successfully sync an FMGo 16 iPhone instance with our FMS16 server-hosted database when all interactions were VPN/LAN based. But in setting up public WAN access, I encountered problems I cannot work around and did not find documentation to address. We opened up our firewall to not require offline users to use VPN. Since we have redundant ISP, I used a hostname as the external address so users could connect over the secondary line if our primary was down. Each of our WAN connections has multiple public IPs, and the public IP assigned to the filemaker server does not match the public ip of the router. When setting up the sync in MirrorSync Configuration java app, an error is reported that "The external sync address you have supplied (fms.company.com) does not match the actual address (x.x.x.154)." I chose to ignore the problem because fms.company.com successfully routes to x.x.x.156--FileMaker Pro, Go, web browsers, etc, have no issue getting to the server. When trying to sync FIleMaker Go, we now have an error and cannot sync: "Communication Error with Server. Error from server: user lacks privilege or object not found: SYNCCLIENT in statement [INSERT INTO SyncClient(timestamp, description, device_id, clone) VALUES (?,?,?,?)] I'm hoping there's a setting that can be adjusted so I can use a domain name for remote WAN syncs.
  10. Thank you Wim. And thank you for emphasizing the need to keep this directory restricted to FMS use–I was surprised that this configuration is now supported since these files should never be touched by anything but the FMS.
  11. I have 2 clients with a similar problem. Each are artist foundations, each about to receive a large volume of digitized video, audio and photography. A portion will be managed by FileMaker, and a portion will simply be stored in a traditional file sharing environment. Neither client can estimate what portion of the 20+TB of expected data will be managed by FileMaker and which by file server. One option is to simply get large RAIDS for both the file server and FileMaker Server, but we were hoping for a more economical and fungible approach to storage. Expandable network storage used by both systems would be ideal conceptually and financially. I understand remotely stored containers are possible now. (FMS16 setup guide says "If you set up additional database or container data folders on remote volumes, the account must also have full permissions to access these remote folders. ") But I have some concerns... Is this approach even advisable from a performance point of view? Must the database reside on the same remote external volume as the containers? Does anyone have a setup like this in the wild?
  12. Ha, you're right there, Comment. Thanks so much for sharing that post. The scripting suggested by the post is relatively simple, but the caveats numerous...record locking, time to execute on thousands of items, modifying record dates...and mapping out a more reliable and efficient way of dealing with those, as you've begun to map out, doesn't look pretty at all. Certainly time to stop working and checkin with the client to see if an internal report with these particular design requirements is worth the effort and time. Thanks again!
  13. Yes, that's it. Thanks for succinctly summarizing. I need to hide layout objects in a leading sub summary part (and shrink that part) if the first record of that part is the first record on a page. The potential problem you raise is a risk I'm willing to test out if one of us could figure a way to get to that point.
  14. I have a report where the subsummary section information needs to also always appear in the header if the subsummary section does not begin at the top of the page. The design problem is when a subsummary starts at the top of a page. The solution: hide either the header or subsummary when the subsummary starts at the top of a page. I found an easy method to hide the header if the subsummary starts at the top, a summary field that restarts counting at subsummary breaks. If the value is 1, hide the header so that only the subsummary appears. HOWEVER, headers don't shrink and the accompanying white space is unacceptable to the client. That lead me down the road of the second option...hide the subsummary elements and shrink that (since these parts do shrink). Because page number is so contextual, all logical calcs I tried failed on me, and I'm unable to write data to the records in preview mode where page numbers are known. Row heights vary, so a simple calc relating to the number of found records won't help. The need to have the same info in the subsummary part and in the header is non-negotiable, the need to not have this duplicate non-negotiable, and since there's no workaround for shrinking the header, I'm focusing on hiding fields and shrinking the subsummary parts.
  15. I need to identify if a record is the first record on a page. I was hoping to use getnthrecord to do so--along with an unstored calc of get (pagenumber) named pageNumberC to determine the page number of the previous record in a report. The formula is getnthrecord ( pageNumberC ; get (record number) - 1). If this value was less than the current record, I would know that the record is the first on a page. It appears that get (pagenumber) is fully contextual and does not function with getnthrecord. No matter which record I select, it only shows the current page number. And as far as I know, FM can't write data while in Preview mode, so I can't store which page a record will fall on in the report. Am I missing something from getnthrecord usage? Any suggestions for accomplishing the goal of determining if a record is the first on a page?
×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use.