The part of your question about what product to buy is off-topic. We call those "shopping questions," and they are almost always off-topic. They are also generally indicative of X/Y problems.
However, there is an underlying problem here that can be addressed. That is allowing your products to drive your process, rather than picking tools that support your process.
Your best bet is to just buy the right tool for the job. In this case, that's probably getting yourself a cork board and some push pins. However, in the unlikely case that the team can't do "the simplest thing that could possibly work," I've also provided some alternative process options for the team to evaluate.
Analysis and Recommendations
Working with Your Current Process
Your real problem (not the X/Y problem you're trying to resolve) is that you want a physical kanban board, but have an unsuitable board that you're trying to manage story cards on anyway.
You have a couple of key options with your current setup:
I can't think of a reason why such solutions wouldn't work, other than that they may gum up the surface of your glass board over time. Even if it does, so what? You can always use adhesive remover when the project is over, or treat it as a delayed cost where the board or its surface need to be repaired or replaced.
However, a better solution would be to define your process and buy the right tools to support that process. You want to use index cards and push pins? Buy a large cork board instead of working around your existing equipment!
If the project doesn't have the budget for a cheap cork board, that's a separate problem for the team to address. An unwillingness or inability to allocate petty cash for basic process support is a much bigger process issue that should be faced head-on.
Inspect-and-Adapt Your Process
You may also prefer to avoid the whole issue entirely. If you can't (or won't) buy the right tools for your process, consider revamping the entire process rather than trying half-measures.
In this case, if you can't replace the glass board with a cork board, or can't find a workable solution with your existing equipment, then you may just want to switch to a different process altogether.
For example, instead of trying to solve the physical board problem, you might consider an electronic kanban board such as Trello or Favro. There are literally hundreds of other clones and work-alikes; some are free, some are commercial, but they all have basic Kanban functionality with additional features and APIs that you'll have to evaluate for yourselves.
You might even consider that a kanban board is a common practice—it's certainly an effective way to visualize the work—but having one is not required by the Scrum framework. While a team must track the work on its Sprint Backlog, that backlog can be a spreadsheet or other artifact just as easily as a kanban board. Perhaps the team would rather have a daily printout of the current burn-down or Sprint Backlog, rather than walking a physical kanban board.
The number of solutions are limited only by your imagination. If you want a physical board, your best bet is to do it right. If you don't actually need a physical board, then have the team develop a supporting process for the daily stand-up that doesn't involve one.