Report ID: 169164383 Can't save web page
Report ID: 169164383
I can't figure out how to add a web page to Zotero.
Steps to reproduce:
1. Start Zotero.
2. Navigate to https://zhidao.baidu.com/question/1737544711358251707.html
4.Clike save web page with snapshot to zotero
Zotero displays the message An error occurred while saving this item. Try again, and if the issue persists see Getting Help for more information.
I can't figure out how to add a web page to Zotero.
Steps to reproduce:
1. Start Zotero.
2. Navigate to https://zhidao.baidu.com/question/1737544711358251707.html
4.Clike save web page with snapshot to zotero
Zotero displays the message An error occurred while saving this item. Try again, and if the issue persists see Getting Help for more information.
Do any pages work? E.g., can you save this page? Could we see a Debug ID from Zotero for that?
if i know how to debug zotero client maybe i can tell you why zotero cause error.
Or,maybe you can send me a debugable zotero client.so i do not need install linux to build from source .....
This is the error you're getting: There's a very good chance this is due to security software on your system interfering with Zotero's file operations, so start by temporarily disabling anything you're running. This isn't an error anyone else has reported, so it's something specific to your system.
-console doesn't change any behavior in Zotero, which is why I said it wouldn't have fixed this. It's also very slow, which is why we don't recommend even using it for debugging anymore. If you find that snapshot saving reliably works when you pass that flag and doesn't work when you don't, it may be due to its making Zotero run slowly, perhaps giving some security software time to perform operations on files. But I'd still be pretty surprised if -console really made a reliable difference.
BTW:i reviewd this error on MAC os 10.12,Centos 7 and windows 7,10 platform. security software ??? may be you can not test on unicode(Like chinese language system) environment....
.........................
(4)(+0000012): SELECT itemID, title FROM items JOIN itemNotes USING (itemID) WHERE libraryID=? AND itemID NOT IN (SELECT itemID FROM itemAttachments) AND itemID IN (2626) [1]
(3)(+0000013): Loaded itemData in 我的文库 in 41 ms
(1)(+0000032): Rolled back DB transaction CHPr3ZwY
(1)(+0000000):
(1)(+0000002): Error: Win error 2 during operation move on file C:\Users\dddang\Zotero\storage\tmp-pHbgZb (系统找不到指定的文件。 ) (No stack trace) From previous event: Zotero.HTTP</this.loadDocuments/onLoad@chrome://zotero/content/xpcom/http.js:940:20
The transaction rollback is in response to the move error, not the other way around. We can only work with the information you give us, so it doesn't make sense to keep sharing new details and then acting shocked that we haven't yet accounted for those.
If you're getting this on another OS, we'd want to see a Debug ID from there too.
no.i did not make symlink.yes.it is a normal folder on all platform.
i do not have only security software on linux or mac system.
Do the same pages fail across all three platforms, and fail consistently on the same computer, or is it more random?
As I said above, your original example URL saves for me on macOS, though saving the snapshot is taking long enough currently that the connector times out after 15 seconds and shows an error. That's just cosmetic, though — the client saves the page fine, though, and doesn't generate the move error.
i save the same page of avove .all of three platforms reported same error.move file cause not found error.i think the system did not create the tmp storage of snapshot case the error.
Basically what should happen here is that Zotero should attempt to save the webpage (using the same saving code as "Save Page As…" in Firefox) to a tmp-* directory in 'storage'. When the save completes, Zotero creates an attachment item, and within that same DB transaction it moves the tmp-* directory to a regular 8-character storage directory. In your case, when it goes to do the move, it's getting an error that the directory doesn't exist. So either the directory isn't being created at all, it's being deleted by something else before it's moved, or files within the directory are in the process of changing (e.g., if the end of the save isn't being properly detected and it's starting to move it too soon) and that somehow shows up as a 'move' error.
Since this is happening for you on three different platforms and hasn't been reported by anyone else, my best guess is that something about your network connection is causing this. I tested the Baidu page above with an artificially slowed-down network connection and it worked fine, but it's possible that if, say, a specific request for one of the page components is blocked or throttled, that could somehow cause problems here. But we may be able to figure out more if you watch what happens in the 'storage' directory during the save.
Yes. the tmp* directory did not create while save snapshot.
Finally,i find what happened for this problem.
If some web page request a resouce that waste much more time,then zotero client will never receive onDocumentReady event,so this error occured....
May be zotero can detect this scene and call stop load page method,just save partial document.