IMD SolutionComputing LibraryShop main store

Troubleshooting

Redirect saved trend files to a new historian server

Edge computing and IIoT Published 6 min read

Short answer

A saved .aaTrend file keeps the server name it was created with, so it hangs when the old historian is gone. Open it in the current Trend client, let the connection attempt finish (observed one to two minutes), save a new copy, then search for the old server name in that copy and replace it with the new one; tag names stay unchanged.

On this page
  1. Why the file still calls the retired server
  2. Prove the replacement server answers first
  3. Repoint the saved trend files
  4. What not to do
  5. Worked example: repointing one 2020.x-era .aaTrend file
  6. Questions asked

A saved trend does not follow its tags to the new historian. Every .aaTrend file stores the server name it was created against, so after the old server is retired the file hangs on open while the connection attempt times out — field reports put that wait at one to two minutes — and the tag list cannot be repointed from inside the trend. The repair is to open the file in the current Trend client anyway, let the attempt finish, save a new copy, and replace the old server name inside that copy. Copies saved by Historian Client 2020.x show unreadable characters in Notepad until that resave; after it, the server name is searchable text. Tag names stay untouched.

Why the file still calls the retired server #

A historical trend holds two separate things: the tags it plots and the Historian server it queries. Keeping the tag names identical on the replacement server makes the migration look trivial, and it is half the work — but nothing reaches into a saved file to update its stored destination. Editing a tag inside the trend changes what is plotted, not where it is read from, which is exactly why the server looks locked while the tags appear fine.

The hang on open is the client working through a connection attempt to a destination that no longer answers. The one-to-two-minute wait is a field observation from systems like this one, not a vendor-specified timeout; the number moves with the client's connection settings and with how the unreachable name resolves or fails to. Wait it out, because the resave that repairs the file format only happens after a completed open.

The unreadable Notepad view on these copies is a format behavior of the client release that wrote them, reported on files saved by Historian Client 2020.x: the old server name is not findable in the jumble, so search-and-replace is dead on that copy. The same file, resaved by the current client, was observed opening as text with the server name searchable. That resave is the hinge of the whole repair.

Prove the replacement server answers first #

Edit no file until one fresh trend proves the new path. In the installed Trend client, create a new trend against the replacement Historian, add one tag you can vouch for end to end — tracing a PLC tag from controller to historian display is the reliable way to pick one — and request an interval you know holds recorded data. Then read the result:

  • Data comes back with sane values and timestamps: the server, the connection and the tag are good, and the fault is the saved binding. Run the repoint procedure below.
  • The connection fails outright: this is a destination, name-resolution, network or authentication fault, and no file edit will fix it; the diagnosis order for historian trend client connection failures applies here first.
  • The tag is missing on the new server: the Historian tag configuration is the next check, not the file.
  • The tag exists but the interval is empty: identical tag names never proved the history moved with them; data availability on the new server is its own question.

Repoint the saved trend files #

  1. Copy every affected .aaTrend file to a working folder and leave the originals untouched; they are the rollback.
  2. Open one copy in the current Trend client and let the connection attempt run to completion — budget one to two minutes, and do not kill the process when the window stops responding.
  3. If the open trend lets you choose its data source, point it at the replacement server and save under a new filename. The client writes its own structure, which makes this the cleanest repair, and the file is done.
  4. If the trend exposes no way to change the server, save it anyway under a new filename and close it.
  5. Open that resaved copy in Notepad and search for the exact old server name.
  6. Replace the occurrences that define the Historian destination with the new server name, leave every tag name alone, and save keeping the file's existing text encoding.
  7. Reopen the edited copy in Trend and inspect every series. One display can carry more than one source reference, so a single replaced occurrence is not proof that every plotted tag now reads the new server.
  8. Request the known interval once more and confirm values and timestamps before you start on the next file.

What not to do #

  • Never save an unreadable file from the text editor: writing the binary content back as text corrupts the structure without changing the connection.
  • Do not kill the client during the hang; the wait is the timeout, and the resave you need only happens after the open completes.
  • Do not run a blanket replace of the server name without looking at each occurrence; a broad replacement can hit labels and unrelated references.
  • Do not overwrite the original file or the unmodified resave; keep both until the edited copy has proven itself.

Worked example: repointing one 2020.x-era .aaTrend file #

Inputs:

  • File: one saved trend with the .aaTrend extension, saved by Historian Client 2020.x — the release reported to produce the unreadable copies.
  • Binding: the file names a retired, unreachable Historian; the replacement server is online and carries identical tag names.
  • Observed open delay with the old server gone: one to two minutes, from field reports — not a documented timeout.

Rule in words: the binding is the stored server name inside the file. The current client's resave rewrites the copy so that name is searchable text; replacing the string redirects the query, and the unchanged tag names stay valid on the new server.

Written out:

  • Open the copy and wait out the connection attempt: one to two minutes, once per file, because each file carries its own binding.
  • Save under a new filename through the client, close it, and open that copy in Notepad.
  • Search the old server name: each occurrence that defines the destination is replaced with the new server name; tag names are left untouched.
  • Reopen in Trend, check every series' source, and pull the known tag over the known interval.

Expected result, on the reported behavior: the resave is the difference — the old server name goes from unfindable in a jumble to searchable text, and after the replacement the trend returns the tag's history from the replacement server.

Decision: if the resaved copy exposes the old server name as searchable text, replace it — that file is repairable, and text replacement is the right route when the interface gives you no server control. If the copy is still unreadable after the resave, stop editing and recreate the trend in the client against the replacement server. And if the fresh test trend never reached the new server at all, the files were never the fault — fix the connection first.

Questions asked

Are .aaTrend files encrypted?

The AVEVA documentation for saving trends states that files with the .aaTrend extension are not encrypted. In practice, copies saved by Historian Client 2020.x have been observed opening in Notepad as unreadable characters, with the server name unfindable, until the newer client resaved them. Treat the resave as the format test that decides which route you use, not as a guaranteed conversion for every release.

Can I change the server inside the Historian Client instead of editing the file?

Yes, where the client exposes it. The documented editor is on the Tools menu: click Servers, select the name in the Server List box, and edit the details for that server. That covers the connections the client itself uses. When an open trend still will not let you repoint its data source, as happens after this kind of server change, the resave-and-replace route on the file is the practical path.

How do I keep new trends from inheriting the old server?

Configure the defaults once in a Default.aaTrend file. The client supports that file as the template for new trend documents, so a replacement server set there is where every new trend starts. Existing files are unaffected; they still carry their own saved server reference and need the repoint individually.

What if the resaved copy is still unreadable in Notepad?

Stop there and recreate the trend. Forcing a text editor to save binary content damages the file structure without changing the connection, so an unreadable resave closes the file-level route for that copy. Build the trend again in the client against the replacement server, using the original file only as the reference for which tags and time ranges it displayed.

How long will the trend hang before it opens?

The observed wait is one to two minutes while the connection attempt to the retired server runs out; it is not the file loading. Wait until the window finishes opening or reports the connection failure, because the resave that fixes the format only happens after a completed open. Killing the process early buys nothing and leaves the copy unconverted.

Keep reading

  • Troubleshooting

    Fix InTouch 2023 Test Connection failure to Historian

    Test Connection from InTouch 2023 to a remote Historian fails for two reasons: the Historian node's firewall blocks the SQL Server listener even though ping works, and the Distributed Name Manager history provider does not support Windows credentials. Prove TCP reachability, then use a SQL Server login stored in Credentials Manager.

    Edge computing and IIoT

  • Comparison

    Operator access to historian trends on night shift

    Put routine historian trend access in front of operators: prebuilt HMI or SCADA trend screens for the recurring questions, point selection over already-collected points for everything else, and a read-only web portal as the alternate route. Keep collection configuration with engineering. No permission setting retrieves samples for a point the historian never collected, so that gap is closed by adding collection, not by access rights.

    Edge computing and IIoT

  • Troubleshooting

    Troubleshooting historian trend client connection failures

    A trend client connection failure has two first checks. If trend pens or Historian Client still load data, compare the failing client's connection definition against a working one instead of restarting the server. If every client fails, verify the historian and SQL services, then test TCP reachability from the failing node to the configured endpoint, port 1433 only if that is the configured port.

    Edge computing and IIoT