For a long time, a fiber cable worked on both allowable wavelengths, and now it does not. Is this a defect of manufacturing? Maybe it's due to weather, other external factors, mechanical damage, or aging? Is it possible to fix?
closed as too broad by Ron Maupin♦ Aug 8 '17 at 20:51
Please edit the question to limit it to a specific problem with enough detail to identify an adequate answer. Avoid asking multiple distinct questions at once. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.
The short answer is that fiber cabling is cheap enough that it is not worth your time or effort to fix a damaged cable in many instances. If you have doubts about a cable, you should just replace it if possible.
In some cases, the fiber is not able to be easily replaced (eg, undersea cables). If, say, you have a long run between buildings, then you might be better off using a OTDR to locate a possible break, and fixing/splicing the cable at that point.
Basically, you have to decide if the effort to fix the cable (if its possible) will be less than running an entire new cable. Most of the time, such as in DC or enterprise environments, it will probably be much cheaper/easier to run a new cable.
As for why the cable no longer works properly....there could be many reasons. Fiber cabling is much less robust than copper cabling. If the cladding or PVC jacket has been worn away, light may be leaking at that point, which may be interfering with the missing wavelength.
Both water intrusion and bending (literally the bend radius of parts of the cable) have detrimental effects that vary with wavelength. As such, a cable disturbance that made a bend tighter would be a classic, textbook reason for losing the longer wavelength - so would a failure of waterproofing. A multiple wavelength OTDR would be a textbook approach to finding that problem. If a section of cable has much higher attenuation at a longer wavelength than a shorter one, you look for a problem at that point in the cable.
Outside the textbook, looking at the more exposed cabling and connections at each end is a good practice, as they are more commonly disturbed, and thus more commonly the problem. Specifics of the fiber you have, the frequencies you had, which one you lost, and your available test equipment and ability to use it would be helpful. If you can relate any changes that were happening to the time that one frequency was lost, that can be helpful - if work was being done in the cable path, or if a network closet or server room had work going on at the time, that helps guide where to look first. It may not have been "network work" - cables and patch cords can be disturbed by electricians, HVAC and plumbers, or even furniture movers and custodial staff, depending on how exposed they are, or where they were routed. Even well-intentioned but ill-informed IT folks can cause a problem.
I think (having built a campus singlemode network from scratch) any decently competent network tech who wants to learn this stuff can, but there are a lot of people who do more harm than good through not understanding what they are dealing with, starting with the importance of clean connectors...and one dirty connector can damage a lot of formerly fine connectors. Depending on the type of "dirt" you won't just make the other connectors dirty - you can harm them to the point that they need to be replaced.
If you have, or can get, an OTDR and are, or can get, or can become someone who knows how to use it, it's the gold standard for finding problems. But you need a fair amount of gold to get one, or even rent one, which is why folks often use lesser tools to try and sort things out. Also, know what you are doing with respect to fibers and eye damage - most fiber lasers are invisible. Best practice is to isolate and disconnect both ends of the fiber before testing/examining, but in any case to use only optical systems that will protect your eye (either with filters, or by inserting a video step) when examining the ends of fibers to be sure they are clean.
If the two wavelengths are being used in opposite directions (as a Bi-Directional single-fiber link) you might also have something simple, like a multi-mode patch-cord on a singlemode system - that is non-blessed in either direction, but might allow reception at its end, while utterly ruining transmission from the end it is at.
I will edit this if you provide more details and I can give a better answer. I am also going to go out on a limb and guess you have lost the lower wavelength and the upper wavelength is still working.
The upper wavelength has less attenuation as it travels through the fiber, so it will arrive at the receiving end with more power. Starting from a point where both wavelengths are functional, if you introduce something into the environment which increases the attenuation, you can reduce the power on the lower wavelength so it is no longer functional.
You have already mentioned some of the causes that can increase attenuation, but I would look first at the fiber patch cords, dirty fiber, or fiber bends as causes. Fiber patch cords are easily damaged and easily replaced. You can easily take the opportunity to clean the fiber properly at this time (get a good non-alcohol based cleaning kit and a fiber scope).
A visual fault indicator (i.e. a bright light source) can also help identify problems quickly in fiber depending on the situation. Anywhere you see the light bleed out would indicate a loss of signal strength and the brighter the light the more loss you are experiencing.
It would be helpful to know more about the fiber... type, location, environment, any recent changes, etc.
Generally speaking a quality cable should last decades if left alone. Patch cables tend to fail due to movement, numerous plug/unplug cycles, and excessive bends. Infrastructure cables (in walls, plenums, underground, or aerial) should last a good long time unless there were material defects in construction or deployment, but age does degrade them, slowly. I'd blamed temperature changes, but I'm not a materials scientist. (something about glass re-crystallization making it brittle and subtly changing it's characteristics.)
As Justin said, if it's an infrastructure cable (read: hard/expensive to replace) find an OTDR to see where and what the problem really is. For all we know, there could be a dirty patch connection somewhere.