IMDTouchKnowledge BaseShop main store

Troubleshooting

Why HMI Programming Software Needs a Ping Reply to Download

HMI and touchscreens Published 6 min read

Short answer

The programming software sends an ICMP echo request to the panel and aborts the download when the reply is blocked, even while remote viewing still works. Ask for one scoped firewall rule: the echo request from your authorized source to the panel address, with its stateful reply. Do not disable the firewall or fake the reply.

On this page
  1. Remote viewing and the download are different paths
  2. What the software does before it will connect
  3. Confirm the address and find where the drop is
  4. The rule to request from the network administrator
  5. Scoping the rule for one programmer at 10.20.5.44 and one panel at 192.168.10.31
  6. Verify and close out
  7. Questions asked

Your HMI still views fine from the office, but the programming software will not download. That combination points at one thing: the software sends an ICMP echo request to the panel's address before it will connect, and when no correct reply comes back it abandons the download, so a firewall dropping the reply breaks every transfer while leaving viewing untouched. The fix is one narrowly scoped rule on the customer's firewall: the echo request from your authorized address to the panel, with the reply allowed back. Disabling the firewall or faking the reply is the wrong repair. Below: how to find where the packet dies, the exact rule to request, and the test that proves it.

Remote viewing and the download are different paths #

A viewing session proves only that its own connection reaches the panel. It says nothing about the reachability check, which runs before any transfer starts. Treat the check as a precondition, not as the transfer itself: once it passes, the project moves over the panel's documented transfer routes, from the software over Ethernet or a USB cable, or by USB memory or SD card. A panel that views perfectly and still refuses a download is not contradicting itself; two different things are being tested, and only one of them is failing.

What the software does before it will connect #

The reported case is an EA9 6-inch panel programmed remotely with C-more 6.12, with earlier versions behaving the same. Per AutomationDirect support, the software (part number EA9-PGMSW) sends a ping to the panel's IP address, and if no correct reply returns it will not connect or download program changes. The reporter confirmed it with an on/off test on the office router: ping block up, download refused; block removed, download completed. Nothing about the panel or the project was wrong, only the reply path. So stop re-checking the project and fix the network path.

Confirm the address and find where the drop is #

Before anyone touches the firewall, gather three facts:

  • The panel's configured address, read from the panel itself or the project. A stale or translated address fails this check honestly. If the address keeps differing between sessions, read where to save C-more panel IP settings in the project.
  • The address the site sees for you: your VPN address if you come in over one, otherwise the public address your traffic leaves from. The rule gets scoped to that.
  • An ICMP test from that same computer and path, with the firewall or router logs to match. A timeout alone does not say where the packet died.

The logs split it three ways, and each way has one fix:

  • The echo request never reaches the site network: the fault is the route or the VPN. No firewall rule at the customer fixes that.
  • The request arrives at the panel and its reply is dropped: the site firewall policy is what to correct.
  • The reply returns to you and the software still refuses: stop changing ICMP. Verify the target address and the transfer settings instead.

The rule to request from the network administrator #

On a stateful firewall, the narrow ask is: permit the echo request from your authorized source to the panel address, and let the session table carry the reply home. That request-and-reply pair is the exchange ICMP, defined in RFC 792, actually uses. Scope both ends to single addresses. If the device cannot track ICMP sessions, the static equivalent is an echo reply from the panel address to your source address only, not open inbound ping. Note the difference: a rule that permits unsolicited echo requests does not necessarily permit the return traffic the software is waiting for.

Administrators filter ICMP for cause; echo traffic is a known abuse vector, and how an ICMP echo flood attack is diagnosed shows what those rules defend against. So ask for the pair, not for ICMP in general.

Two shortcuts to refuse, stated once: do not ask a customer to disable their firewall for a programming session, and do not put another box on the network to answer the ping. A forged reply proves nothing about the panel and buries a real fault such as a duplicate or mistranslated address. There is also no documented setting that turns the check off in the software; if you need that behavior changed, take it to the vendor rather than an undocumented switch.

Scoping the rule for one programmer at 10.20.5.44 and one panel at 192.168.10.31 #

Inputs: an EA9 panel at 192.168.10.31, mask 255.255.255.0 (/24); a programming computer at 10.20.5.44 reaching the site through a VPN; a stateful, default-deny firewall between them. The panel, the software version and the ping behavior come from the reported case; the two addresses are illustrative RFC 1918 private values, the normal situation over a VPN, so substitute the panel address and source address you recorded above.

The rule in words: permit one ICMP echo request from 10.20.5.44 to 192.168.10.31 and let the firewall's session table return the echo reply. Nothing wider.

The arithmetic of why the scoping matters:

  • Usable addresses in the panel's /24: 2^(32-24) - 2 = 2^8 - 2 = 256 - 2 = 254.
  • A rule that allows ping to the whole panel VLAN names 254 destinations. The pair rule names 1.
  • 254 / 1 = 254, so the pair rule exposes 1/254 of that destination space: 1 / 254 = 0.0039, about 0.4%.

The pass test, run from the programming computer: ping the panel and read the summary. Sent = 4, Received = 4, Lost = 0 (0% loss), and 4 - 4 = 0 packets lost is the pass line; anything else is still a fault.

Decision: hold the rule at one source address and one destination address. That is the whole rule, and anything wider is a concession, not a requirement. If the pair rule is in and the reply still does not return, the drop is upstream, so work the route or the VPN. If the reply returns and the software still refuses, the fault is addressing or transfer settings, not ICMP.

Verify and close out #

  1. From the same computer and remote path that will do the programming, repeat the ping test and confirm the 0% loss pass line.
  2. Start the download from the software; the connection check passes and the transfer begins.
  3. Let the transfer finish and confirm the panel runs the updated project.
  4. If the change was a time-limited one, remove it and confirm the permanent policy still covers this workflow.
  5. If ping succeeds and the download still fails, ICMP is done; the fault is in the transfer itself, such as the panel-mode case in why an HMI transfer fails in initial start mode despite ping.

Questions asked

Why does remote viewing work while the programming download fails?

Viewing rides a connection the site already permits, while the download attempt starts with a separate ICMP echo check that many policies treat on its own. Passing one proves nothing about the other. Confirm the echo reply actually returns to the programming computer before you blame the transfer.

Can the ping requirement be turned off in the software?

No supported switch for it is documented. The behavior was reported on version 6.12 and earlier, and the vendor's support described the check as part of connecting. The dependable fix is the scoped firewall rule; if you need the check itself changed, get a supported answer from the vendor instead of trusting an undocumented setting.

Is allowing inbound ping from the internet the same fix?

No. The packet the software waits for is the panel's reply to its own request. A rule permitting unsolicited echo requests from anywhere is far broader than that return path, and it still does not guarantee the reply leaves the site. Ask for the request plus stateful reply, scoped to your source address and the panel address.

Can another computer or the router answer the ping for the panel?

No. A forged reply proves the network can answer, not that the panel is reachable, and it buries a real fault such as a duplicate or mistranslated address. If the genuine reply will not return, work the route or the firewall with the site administrator instead.

What if ping now succeeds but the download still fails?

Then ICMP is no longer the fault, and more firewall changes will not help. Recheck the target address in the project, the panel's transfer mode, and the transfer settings; a panel waiting in a transfer or initial start mode can refuse a project even with perfect reachability.

Keep reading

  • Troubleshooting

    EcoStruxure Operator Terminal Expert download fails to HMI

    A failed Operator Terminal Expert download has three separate causes: Windows firewall blocking the transfer executable, fixable with inbound and outbound allow rules scoped to EcoStruxure Operator Terminal Expert.exe; a VMware virtual USB controller that needs its compatibility set to USB 1.1; or a terminal still running Vijeo Designer firmware, which no Operator Terminal Expert project can load until a manufacturer recovery converts it. Match the fix to the runtime you read on the terminal.

    HMI and touchscreens

  • Troubleshooting

    Fixing a PCS 7 OS download that fails with SQL errors

    The transfer fails because the target-side WinCC SQL databases, the CC_* copies, stay attached to SQL Server or sit in RECOVERY_PENDING, so WinCCOM cannot replace them. Close Runtime and all editors, reboot the target, remove the obsolete CC_* databases through the SQL administration layer rather than by deleting .mdf files, delete the stale CC_OM DSN, then run one clean download.

    HMI and touchscreens

  • Troubleshooting

    Fixing an HMI Project Upload That Reports Communication Failed

    Treat a failed HMI project upload as three separate branches: the installer's parallel-port dongle driver message is a licensing fault, a "communication failed" report before any progress is a transfer-path mismatch, and a "file system error" after the panel answers is a storage or destination fault. Identify the layer from the message and the moment it appears, fix that layer only, and retest. A dongle fault never blocks the cable, and a successful ping never proves the transfer service is up.

    HMI and touchscreens