I have an instance of accelerated attack in bubbler blocks in a furnace making green glass, resulting in a breakout.  I have been told that this is unusual.  That normally as the hole around the bubbler tubes gets deeper, the bubbler will stop working before a breakout because of the accumulation of low MP constituents from the AZS refractory.  The breakout occurred after two years of service, instead of the 4+ years expected.  Any comments?